Navigation using two-dimensional map data for autonomous systems and applications

Abstract

The invention relates to navigation using two-dimensional map data for autonomous systems and applications. Embodiments of the present disclosure relate to a system and method for locating one or more systems using 2D map data. The method may include determining an image position of a representation of a portion of an object in an image corresponding to an environment. In some embodiments, the method may also include determining one or more predicted image positions corresponding to the image position of the representation of the portion of the object. The method may also include comparing the one or more horizon positions of the portion of the object to the one or more predicted image positions, and determining a cost based at least on the comparison between the one or more horizon positions and the one or more predicted image positions. In addition, the method may include locating the system to the 2D map data based on the determined cost.

Description

Background Art

[0001] Mapping, localization, and navigation are important processes for many machines, particularly for autonomous and semi-autonomous machines (e.g., ego-machines). Map data corresponding to a map, sensor perception, or a combination thereof is typically used to localize the ego-machine relative to the map in order to make planning and control decisions. In some cases, localization processes associated with the ego-machine may be performed using sensor data generated using one or more sensors and map data corresponding to a map of an area associated with an environment in which the ego-machine may be located. One or more localization processes may compare the sensor data generated using the one or more sensors with map data corresponding to the map. In addition, the localization process may update a pose corresponding to the ego-machine based on a comparison between the sensor data and the map data corresponding to a map of the environment.

[0002] In a typical deployment scenario, the map data corresponding to the environmental map may include three-dimensional (3D) map data. Accordingly, the positioning process may include one or more processes by which one or more postures corresponding to the machine can be determined in the environment. However, in some cases, 3D map data may not be available for the machine or is not feasible for use with the machine. For example, because regulations limit what data can be collected and used by, for example, the machine, etc., the map data corresponding to the Chinese environmental map may be only two-dimensional (2D). As an additional example, the 3D map data may contain more map data than the map data that can be processed by one or more systems using the machine and / or one or more other systems corresponding to the machine. Therefore, it may be impossible to use 3D map data, or it may be too expensive.

[0003] To address these issues and / or potential positioning issues, 2D data can be used in one or more positioning processes without being affected by regulations, processing availability, and / or other limitations associated with the present machine. For example, one or more conventional positioning techniques using 2D data can use multiple cameras to generate image data corresponding to multiple fields of view at multiple corresponding time stamps. Continuing with this example, since one or more conventional positioning techniques include multiple cameras and multiple time stamps, the positioning determination can also include triangulation of specific feature poses corresponding to the object over time. Triangulation of feature poses over time can be performed to build a map of the corresponding environment; for example, using one or more positioning and mapping techniques, such as simultaneous localization and mapping (SLAM). An example limitation of these conventional techniques is that any error associated with the triangulation of a specific feature pose may affect the positioning result. In addition, in many cases, errors corresponding to the triangulation of feature poses over time may propagate, thereby exacerbating errors corresponding to the accuracy of the positioning determination over time. The propagation of errors over time may cause the positioning process to be invalid, inaccurate, and / or unusable. Summary of the invention

[0004] According to one or more embodiments of the present disclosure, the image position of a portion of an object in an image can be determined, wherein the image can correspond to a specific environment. In some embodiments, the image position can be defined in a 2D image space based on image data corresponding to the image (e.g., image data generated using a camera). In some embodiments, one or more predicted image positions corresponding to the image position of the representation of the object can be determined based on one or more different considerations. These considerations include, for example, a ground representation in a real-world environment; a vanishing point that can be projected into the image space, and / or the image position of a portion of the representation of the object. In some embodiments, the ground plane position of a portion of the object can be compared with the predicted image position. The ground plane position can be determined using a 2D map of the environment projected onto the 2D image space. Additionally or alternatively, the cost can be determined based on a comparison between the ground plane position and the predicted image position. In addition, in some embodiments, the system can be positioned based on the determined cost.

[0005] Embodiments of the present disclosure may allow for positioning without errors corresponding to triangulating feature poses corresponding to one or more objects over time. By determining one or more predicted image locations corresponding to a portion of an object and comparing the one or more predicted image locations to one or more ground plane locations determined using a 2D map projected in a 2D image space, one or more positioning determinations may be performed more accurately and using less data than one or more conventional positioning techniques. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The present system and method for navigating an autonomous or semi-autonomous machine using 2D map data will be described in detail below with reference to the accompanying drawings, wherein:

[0007] Figure 1A A flowchart of a positioning process according to one or more embodiments of the present disclosure is shown;

[0008] Figure 1B An example diagram representing an image according to one or more embodiments of the present disclosure is shown;

[0009] Figure 2A is a diagram representing an example environment associated with determining and / or generating positioning data using a positioning system according to one or more embodiments of the present disclosure;

[0010] Figure 2B An example diagram of a representation image according to one or more embodiments of the present disclosure is shown; one or more first predicted image positions and one or more second predicted image positions corresponding to an object;

[0011] Figure 2C-2D shows example diagrams respectively representing an image and a vanishing point that may be projected into the image space according to one or more embodiments of the present disclosure;

[0012] Figure 2E-2G shows example diagrams respectively representing an image and one or more predicted image positions corresponding to a vanishing point according to one or more embodiments of the present disclosure;

[0013] Figure 3 An example graph showing a cost comparison between one or more predicted image locations and one or more sets of projected 2D map data corresponding to various points of interest associated with an object according to one or more embodiments of the present disclosure is shown;

[0014] Figure 4 is a flow chart illustrating a method of positioning a system using 2D map data according to one or more embodiments of the present disclosure;

[0015] Figure 5A is an illustration of an example autonomous vehicle according to one or more embodiments of the present disclosure;

[0016] Figure 5B According to one or more embodiments of the present disclosure Figure 5A Examples of autonomous vehicle camera positions and fields of view;

[0017] Figure 5C According to one or more embodiments of the present disclosure Figure 5Aa block diagram of an example system architecture for an example autonomous vehicle;

[0018] Figure 5D A cloud-based server and a Figure 5A System diagram of an example of communication between autonomous vehicles;

[0019] Figure 6 is a block diagram of an example computing device suitable for implementing one or more embodiments of the present disclosure; and

[0020] Figure 7 is a block diagram of an example data center suitable for implementing one or more embodiments of the present disclosure. DETAILED DESCRIPTION

[0021] One or more embodiments disclosed herein may involve one or more processes for positioning one or more systems (including autonomous and / or semi-autonomous systems (e.g., one or more host machines)) using 2D map data, wherein the 2D map data may correspond to an environment in which one or more host machines may be located. In the present disclosure, references to map building and positioning processes may refer to processes by which a pose (e.g., position and / or orientation) corresponding to the host machine and / or sensors corresponding to the host machine may be determined. In some embodiments, sensor data corresponding to the host machine may be generated using one or more image sensors (e.g., cameras), which may be configured to generate one or more images corresponding to the environment in which the host machine may be located. In some embodiments, the image data corresponding to the environment may include a 2D representation of one or more objects corresponding to the environment. The one or more objects may include one or more objects in the environment that may be included in the map data, such as, for example, one or more street signs, poles (telephone poles, electric poles, etc.), street lights, trees, medians, fences, ramps, and other objects that may be included in the map data corresponding to the map of the environment.

[0022] In some embodiments, the image data corresponding to a particular image may correspond to a particular image space. The particular image space may include a mathematical representation for representing and / or organizing data in a 2D environment. In some embodiments, a particular coordinate system utilizing two spatial dimensions may be used to define the particular image space. Additionally or alternatively, the image space may be defined by a particular image.

[0023] In some embodiments, a specific image corresponding to the image space may include a representation of one or more objects corresponding to the environment. In addition, a specific image corresponding to the image space may include a representation of the "ground", which may include a representation of the ground in the real world.

[0024] In some embodiments, the map corresponding to the map data may be a 2D representation of an environment that may be generated using the map data. The environment may correspond to a location where the machine and / or image sensor may be located. For example, the 2D map data may correspond to one or more objects, locations, streets, etc., without data and / or information corresponding to the height of one or more objects.

[0025] In some embodiments, map data corresponding to a map of the environment may be compared to 2D image data associated with the environment. In some embodiments, comparing the map data and image data corresponding to the map may include comparing posture data corresponding to respective postures of one or more landmarks that may be shared between the image data and the map data. In some embodiments, the machine and / or sensors corresponding to the machine may be positioned based on the comparison between the map data and the image data. Embodiments of the present disclosure may allow one or more systems corresponding to the environment to be positioned using 2D map data; and in some embodiments, using a single monocular camera. In some embodiments, one or more embodiments may help improve the effectiveness of one or more systems by reducing the number of errors corresponding to one or more positioning determinations.

[0026] For example, one or more embodiments may allow for positioning without the errors associated with triangulating feature poses corresponding to one or more objects over time (e.g., one or more SLAM techniques). By determining one or more predicted image locations corresponding to a portion of an object and comparing the one or more predicted image locations to one or more ground plane locations determined using a 2D map projected in a 2D image space, one or more positioning determinations may be performed more accurately and using less data than one or more conventional positioning techniques.

[0027] One or more embodiments disclosed herein may involve one or more processes of locating one or more systems (including autonomous and / or semi-autonomous systems) using 2D map data, where the map data may correspond to an environment in which one or more present machines may be located. One or more present machines disclosed herein may include any applicable machine or system capable of performing one or more autonomous and / or semi-autonomous operations. Example present machines may include, but are not limited to, vehicles (land, sea, space, and / or air), robots, robotic platforms, etc. For example, the present machine computing applications may include one or more applications executable by an autonomous vehicle or semi-autonomous vehicle, such as regarding Figures 5A-5DAn example autonomous or semi-autonomous vehicle or machine 500 (also referred to herein as "vehicle 500" or "present machine 500") is described. In this disclosure, references to an "autonomous vehicle" or "semi-autonomous vehicle" may include any vehicle that can be configured to perform one or more autonomous or semi-autonomous navigation or driving operations. Thus, such vehicles may also include vehicles that require an operator or an operator may also perform such operations.

[0028] The systems and methods described herein may be used by, but are not limited to, non-autonomous vehicles or machines, semi-autonomous vehicles or machines (e.g., in one or more adaptive driver assistance systems (ADAS)), autonomous vehicles or machines, manned and unmanned robots or robotic platforms, warehouse vehicles, off-road vehicles, vehicles coupled to one or more trailers, spacecraft, boats, shuttles, emergency response vehicles, motorcycles, electric or motorized bicycles, aircraft, engineering vehicles, submarines, drones, and / or other vehicle types. In addition, the systems and methods described herein may be used for a variety of purposes, such as, but not limited to, machine control, machine motion, machine driving, synthetic data generation, model training, perception, augmented reality, virtual reality, mixed reality, robotics, safety and surveillance, simulation and digital twins, autonomous or semi-autonomous machine applications, deep learning, environmental simulation, object or actor simulation and / or digital twins, data center processing, conversational AI, light transport simulation (e.g., ray tracing, path tracing, etc.), collaborative content creation of 3D assets, cloud computing, and / or any other suitable application.

[0029] Embodiments of the present disclosure include in a variety of different systems, such as automotive systems (e.g., control systems for autonomous or semi-autonomous machines, perception systems for autonomous or semi-autonomous machines), systems implemented using robots, aviation systems, medical systems, boating systems, smart area monitoring systems, systems for performing deep learning operations, systems for performing simulation operations, systems for performing digital twin operations, systems implemented using edge devices, systems including one or more virtual machines (VMs), systems for performing synthetic data generation operations, systems implemented at least in part in a data center, systems for performing conversational AI operations, systems implementing one or more language models (e.g., one or more large language models (LLMs) that process text, audio, images, sensors, and / or other data types to generate one or more outputs), systems for hosting real-time streaming applications, systems for presenting one or more of virtual reality content, augmented reality content, or mixed reality content, systems for performing light transport simulations, systems for performing collaborative content creation of 3D assets, systems for performing generative AI operations, systems implemented at least in part using cloud computing resources, and / or other types of systems.

[0030] These and other embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the drawings are diagrams and schematics of such example embodiments and are not restrictive and are not necessarily drawn to scale. In the drawings, unless otherwise specified, features with the same number represent the same structure and function.

[0031] refer to Figure 1A , Figure 1A A flowchart 100 for a positioning process according to one or more embodiments of the present disclosure is shown. In some embodiments, a positioning system 104 may be used to generate, calculate, and / or otherwise determine positioning data 110 corresponding to one or more machines and / or sensors corresponding to one or more machines using image data 106 generated using an image sensor 102 and 2D map data 108 corresponding to a 2D map.

[0032] In some embodiments, the image sensor 102 may include one or more types of image sensors, such as a charge coupled device (“CCD”), a complementary metal oxide semiconductor (“CMOS”) sensor, and the like. In some embodiments, CCD and CMOS sensors may be used with one or more cameras to capture and / or generate image data 106 corresponding to one or more captured images. For example, the image sensor 102 may be used with a monocular camera to generate image data 106 corresponding to an image of an environment in which the system may be located. In some embodiments, the image sensor 102 generating the image data 106 may be an example of a plurality of image sensors 102 that may be configured to capture image data 106 contained in one or more images corresponding to an environment in which each image sensor 102 may be located. In some embodiments, the image sensor 102 may be located in the same environment as the positioning system 104.

[0033] In some embodiments, the image sensor 102 may be configured to generate image data 106 associated with one or more images corresponding to the environment in which the image sensor 102 is located. In some embodiments, the image sensor 102 may also be configured to generate other data corresponding to the image. For example, the image sensor 102 may generate metadata corresponding to the image—e.g., a timestamp, sensor type, location information, etc. In some embodiments, the image sensor 102 may be used in conjunction with one or more imaging systems (e.g., cameras), which may use the image data 106 to generate one or more images.

[0034] Image data 106 may include data and / or information corresponding to an image that may be captured that corresponds to an environment and / or a portion of an environment in which image sensor 102 may be located. In some embodiments, the image corresponding to image data 106 may be a 2D visual representation of the environment or portion of an environment corresponding to image data 206.

[0035] In some embodiments, the image data 106 corresponding to a particular image may correspond to a particular image space. The particular image space may use a particular coordinate system to mathematically represent the location of one or more objects and / or portions of the image 120. The particular coordinate system may include two axes, which may include an x-axis and a y-axis, as shown in the image 120, respectively. i and i In some embodiments, a reference to one or more locations in image space corresponding to one or more portions of an image may include a reference to a pair of coordinates, such as (x i ,y i ).

[0036] In some embodiments, a specific image corresponding to the image space may include a representation of one or more objects corresponding to the environment. In addition, a specific image corresponding to the image space may include a representation of the "ground", which may include a representation of the ground in the real world.

[0037] Reference Figure 1B Example representations of image data 106 corresponding to particular images are described and / or shown. Figure 1B An example diagram representing an image 120 according to one or more embodiments of the present disclosure is shown. Although shown in the context of a vehicle on a road, the image 120 may include 2D representations of multiple different environments corresponding to different sensors - for example, referring to Figure 1A The image sensor 102 is described, and / or one or more machines and / or systems corresponding to different sensors.

[0038] In some embodiments, image 120 may be captured and / or generated using an image sensor, such as, for example, image sensor 102. Figure 1B As shown, image 120 may be captured using, for example, a monocular camera or other imaging system located in the middle of the vehicle. Figure 1B As shown, a portion of a front hood and / or windshield may be displayed corresponding to vehicle 118. Image 120 shows a perspective corresponding to an image sensor associated with vehicle 118 at a particular time and corresponding to a particular location. In some embodiments, image 120 may also establish a particular angle of the image sensor.

[0039] In some embodiments, the 2D representation of the environment may be a representation of a particular image space that mathematically represents the position of one or more objects and / or portions of image 120 relative to origin 114. In some embodiments, origin 114 may define and / or establish a reference point to which one or more portions of image space may be related. In some embodiments, origin 114 may be defined by an x-axis and a y-axis, as shown in image 120 as x and y, respectively. i and i , where the origin can be defined as corresponding to the reference point (0,0) in image space.

[0040] In some embodiments, from the origin 114, one or more locations of an object (e.g., object 112) may be defined. In some embodiments, the origin 114 may be located at the upper left corner of the image 120. Additionally or alternatively, the origin 114 may be located and / or defined at one or more other locations in the image 120, for example, the lower right corner, the lower left corner, the upper right corner, the lower center corner of the image 120, and any other point in the image 120. In some embodiments, the origin 114 may be established and / or defined corresponding to, for example, each image 120 generated using the image sensor 102. For example, the image 120 may be a first image representing a first image space having a first origin 114. Continuing with this example, another image may be obtained from the same environment using a different image sensor, a different angle, a different time, etc. The second image may represent a second image space having a second origin 114. In some embodiments, the image space represented using the image 120 may be limited by the image 120 and / or may be different for each image generated using the image sensor (e.g., the image sensor 102).

[0041] In some embodiments, image 120 may include one or more objects and / or landmarks, such as object 112. Object 112 in image 120 may be a 2D representation of a sign; specifically, a speed limit sign corresponds to a section of road where vehicle 118 may be traveling. In some embodiments, object 112 may correspond to a 2D representation of a vehicle 118 using coordinates (x i ,y i ) is a specific portion of an environment defined by an image 120. As shown in image 120, from the perspective of vehicle 118, object 112 in real-world space may be located to the right of vehicle 118 and at a certain distance in front of vehicle 118.

[0042] In some embodiments, one or more portions of object 112 may be identified corresponding to one or more corresponding locations in image space. For example, object 112 may include a first side 112a that may represent a right portion of object 112. The object may also include a second side 112b that may represent a left portion of object 112. In some embodiments, a footprint of object 112 or a size corresponding to object 112 may be determined in image space corresponding to image 120 using coordinates corresponding to one or more portions of object 112 (e.g., first portion 112a and / or second portion 112b).

[0043] In some embodiments, image 120 may include horizon 116. Horizon 116 may indicate a portion of image 120 that may represent a ground plane corresponding to a real-world environment. For example, horizon 116 may be a dividing point or dividing line between a representation of the ground in a real-world environment and a representation of the sky or above the ground in the real-world environment. The real-world environment may be defined in three dimensions (X, Y, Z), and a ground plane corresponding to the ground may be defined, for example, using a plane (X, Z). In an image space corresponding to image 120, horizon 116 may indicate a ground representation that may correspond to a ground plane (X, Z) in a real-world environment.

[0044] Can Figure 1B Modifications, additions, or omissions may be made without departing from the scope of the present disclosure. For example, the placement of the origin 114, the environment captured by the image 120, the angle of the image sensor capturing the image 120, the number of objects 112 and / or the number of parts of the objects 112, the placement or presence of the horizon 116, the presence of the vehicle 118, etc. The specifics presented and discussed are intended to aid in explaining and understanding the concepts of the present disclosure and are not intended to be limiting.

[0045] Back to Figure 1A Description, except Figure 1B In addition to the image 120 described and shown in FIG. 1 , the image data 106 may also include metadata corresponding to the image sensor 102 and / or the image 120. In some embodiments, the metadata may indicate one or more inherent characteristics associated with the image sensor 102. For example, the metadata may include the time of day corresponding to when the image data 106 was captured and / or generated, the type of image sensor 102, the resolution corresponding to the image, the lens type, focal length, aperture, lens quality, shutter speed, image processing, etc. In some embodiments, the metadata that may be included in the image data 106 may be generated as a result of one or more calibration processes.

[0046] Additionally, in some embodiments, image data 106 corresponding to an image (e.g., image 120) can be transmitted, sent, or otherwise communicated to positioning system 104. In some embodiments, positioning system 104 can correspond to the same system and / or machine as image sensor 102. For example, image sensor 102 can be located in the machine. Continuing with this example, positioning system 104 can also be located in the machine, and sensor data 106 can be communicated between image sensor 102 and positioning system 104 using one or more cables and corresponding data transmission protocols.

[0047] Additionally or alternatively, image sensor 102 may be associated with one or more other systems and / or machines than the system and / or machine to which positioning system 104 may correspond. For example, image sensor 102 may correspond to the present machine, also in the context of the present machine navigating a particular environment. For example, an image (e.g., image 120) may be captured using image sensor 102, which may be located on or within the present machine. Continuing with this example, image data 106 generated using image sensor 102 may be sent to one or more other systems, servers, edge servers, computing systems, etc., to which positioning system 104 may correspond.

[0048] In some embodiments, the positioning system 104 may be configured to perform positioning based on a comparison between the image data 106 and 2D map data 108 corresponding to the 2D map. In some embodiments, the positioning system 104 may be configured to receive and / or otherwise acquire the 2D map data 108.

[0049] In some embodiments, the 2D map may be a 2D representation of an environment that can be generated using the 2D map data 108. The environment may correspond to the location where the image sensor 102 is located. For example, the 2D map data 108 may correspond to one or more objects, locations, streets, etc., without data and / or information corresponding to the height of one or more objects. For example, a 2D map may include data and / or information associated with a 2D representation of a real-world space. For example, the real-world space may be defined using a Cartesian coordinate system. More specifically, the real-world space may be defined using three dimensions (X, Y, Z), where the XZ definition corresponds to a plane on the ground. In some embodiments, the 2D map data may include data corresponding to a 3D environment that may be flattened and defined using an XZ ground plane. Therefore, the coordinates corresponding to an object represented using the 2D map data 108 may include coordinates in the form of (X, Z).

[0050] In some embodiments, the 2D map data 108 may show one or more locations corresponding to one or more objects and / or features of a particular environment. In some embodiments, the 2D map data 108 corresponding to the 2D map may include information corresponding to one or more locations of one or more objects. For example, the 2D map data 108 may include coordinate and location information determined using data from one or more sources, such as, for example, one or more global information systems (GIS) databases, map building services, aerial images, etc. In these and other embodiments, the 2D map may show one or more objects in a manner that may correspond to the location information corresponding to the one or more objects included in the 2D map data 108.

[0051] For example, the pole may be located in real-world space (e.g., a telephone pole, a light pole, etc.). The real-world space may be defined using a coordinate system (X, Y, Z). In some embodiments, X defines the lateral or horizontal direction and the distance from the real-world origin to the pole, Z defines the depth from the real-world origin to the pole, and Y defines the height or vertical direction and the distance from the real-world origin to the pole. In some embodiments, the real-world origin may be different from the origin 114 associated with one or more image spaces corresponding to one or more individual images (e.g., image 120). Continuing with this example, the 2D map data 108 may correspond to the position of the pole relative to the origin on an XZ plane defined in the real-world space - the XZ plane is referred to as the "ground plane" hereinafter. In some embodiments, the 2D map may be a visual representation of the 2D ground plane.

[0052] As an additional example, the 2D map data 108 may correspond to a map further described in the present disclosure, such as for example Figure 1B A 2D map of the same environment represented by image 120 is described and / or shown. 2D map data 108 may include, for example, locations of objects 112, roads depicted in image 120, and the like.

[0053] In some embodiments, positioning system 104 may be configured to receive and / or otherwise acquire 2D map data 108 , a 2D map corresponding thereto, and / or image data 106 to determine, calculate, and / or otherwise generate positioning data 110 .

[0054] In some embodiments, the positioning system 104 may include code and routines configured to allow a computing system to perform one or more operations. Additionally or alternatively, the positioning system 104 may be implemented using hardware, including one or more processors, CPU graphics processing units (GPUs), data processing units (DPUs), parallel processing units (PPUs), microprocessors (e.g., for performing one or more operations or controlling the execution of one or more operations), field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), accelerators (e.g., deep learning accelerators (DLAs)), and / or other processor types. In these and other embodiments, the positioning system 104 may be implemented using a combination of hardware and software. In the present disclosure, operations described as being performed by the positioning system 104 may include operations that the positioning system 108 may instruct the corresponding computing system to perform. In these or other embodiments, the positioning system 104 may be implemented by one or more computing devices, such as with respect to Figure 5A –5D, Figure 6 and / or Figure 7 A computing device is described in more detail.

[0055] In some embodiments, the positioning system 104 may compare the image data 106 and the 2D map data 108 by projecting, superimposing, or otherwise integrating the 2D map in the image corresponding to the image data 106. In some embodiments, comparing the 2D map data 108 and the image data 106 may include comparing pose data corresponding to respective poses of one or more objects that may be common between the image data 106 and the 2D map data 108. In some embodiments, the pose data corresponding to the one or more objects in the 2D map data 108 corresponding to the 2D map may be determined based on a corresponding pose (e.g., position and orientation) of the host machine and / or the image sensor corresponding to the image data 106. For example, the image data 106 may include a depiction of a road sign contained in the image, and the 2D map data 108 may include corresponding pose data indicating a ground plane position corresponding to the road sign based on position and orientation information associated with the host machine.

[0056] In some embodiments, pose data indicating the ground plane position of the road sign may be projected or superimposed onto the image space. Furthermore, the projected pose data indicating the ground plane position of the road sign may be compared with a depiction of the road sign contained in the image - for example, the present disclosure further discloses, for example, Figure 1B An image 120 is described and / or shown.

[0057] In some embodiments, the positioning system 104 may be configured to determine a cost or error based on a comparison between the projected pose data indicating the ground plane position of the road sign and the depiction of the road sign contained in the image corresponding to the image data 106. In some embodiments, a cost or error that meets a certain threshold may indicate that the pose data corresponding to the 2D map data 108 matches or nearly matches the depiction of the road sign contained in the image corresponding to the image data 106. Accordingly, a cost or error that meets a certain threshold may indicate that the estimated pose data associated with the image sensor 102 may be correct or nearly correct. In some embodiments, correct or nearly correct may indicate that the position and / or orientation of the image sensor 102 in the 2D map data 108 corresponding to the 2D map matches or nearly matches the image data 106 corresponding to the image.

[0058] Additionally or alternatively, a cost or error that does not meet a particular threshold may indicate that the 2D map data 108 corresponding to the pose associated with the image sensor 102 does not match the depiction of road signs contained in the image, and accordingly, the position and / or orientation of the image sensor 102 in the 2D map does not match the image data 106 corresponding to the image. In some embodiments, a 2D map that does not match the image data 106 may indicate that the position and / or orientation of the image sensor 102 in the 2D map data 108 is incorrect.

[0059] In some embodiments, in response to the cost and / or error not meeting a particular threshold, the positioning system 104 may be configured to adjust the estimated pose of the machine corresponding to the image sensor 102 and / or associated with the image sensor 102. In some embodiments, the estimated pose may be iteratively adjusted until the cost meets the particular threshold. In some embodiments, the cost meeting the particular threshold may indicate that the pose corresponding to the image sensor 102 is correct relative to the objects in the environment. In some embodiments, the comparison of the image data 106 corresponding to the image with the 2D map may be described in the present disclosure, for example, with respect to Figure 2A-3 Described and / or shown in greater detail.

[0060] In some embodiments, the positioning system 104 may be configured to generate positioning data 110 based on a comparison between the image data 106 and the 2D map 108. The positioning data 110 may include data and / or information corresponding to one or more locations corresponding to the image sensor 102 and / or one or more systems, machines, etc. that may correspond to the image sensor 102. In some embodiments, the positioning data may include position data (e.g., latitude and longitude, position data corresponding to the 2D map 108, coordinates associated with one or more other maps or coordinate systems, etc.), orientation data (e.g., rotation information corresponding to one or more different axes, such as roll, pitch, and yaw data corresponding to rotations about three different axes), cost data or other error data associated with the location of the image sensor 102 and / or the corresponding system, position data corresponding to one or more other objects and / or landmarks in the environment, timestamp data, etc.

[0061] In some embodiments, the positioning data 110 may be used by one or more systems, for example, to generate one or more control commands, etc. For example, one or more planning systems and / or subsystems corresponding to the present machine may use the positioning data 110 to generate one or more planned paths from a first location to a second location in an environment. Additionally or alternatively, the positioning data 110 may be used by one or more perception systems and / or subsystems to determine whether one or more evasive actions need to be taken to avoid one or more objects, obstacles, pedestrians, etc.

[0062] Can Figure 1A Modifications, additions, or omissions may be made without departing from the scope of the present disclosure. For example, the amount of image data 106, the environment corresponding to the image data 106 and / or the 2D map 108, the number of systems, subsystems, modules, etc. for performing operations corresponding to the positioning system 104, the number of image sensors, and the amount and type of positioning data 110 may vary. The specifics presented and discussed are helpful in explaining and understanding the concepts of the present disclosure and are not meant to be limiting.

[0063] Figure 2A 2 is a flow chart of determining and / or generating positioning data 210 using positioning system 204 according to one or more embodiments of the present disclosure. In some embodiments, example environment 200 may be an example implementation of environment 100, and positioning system 204 may be, for example, an example implementation of the present disclosure. Figure 1AAn example implementation of the positioning system 104 is further described and / or illustrated. In some embodiments, the positioning system 204 can be an example implementation of one or more other positioning systems, such as described in U.S. Non-Provisional Application No. 17 / 007,873, filed on August 31, 2020, entitled “MAP CREATION AND LOCALIZATION FOR AUTONOMOUS DRIVING APPLICATIONS,” which is incorporated herein by reference in its entirety.

[0064] In some embodiments, the positioning system 204 may include one or more DNNs 202 , comparison modules 216 , and / or positioning modules 270 , which may be configured to calculate, generate, and / or otherwise determine positioning data 210 based on the image data 206 and the 2D map data 208 .

[0065] In some embodiments, the DNN 202, the cost space system 216, and / or the positioning module 270 may include code and routines configured to allow a computing system to perform one or more operations. Additionally or alternatively, the DNN 202, the comparison module 216, and / or the positioning module 270 may be implemented using hardware, including one or more processors, CPU graphics processing units (GPUs), data processing units (DPUs), parallel processing units (PPUs), microprocessors (e.g., for performing one or more operations or controlling the execution of one or more operations), field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), accelerators (e.g., deep learning accelerators (DLAs)), and / or other processor types. In these and other embodiments, the DNN 202, the comparison module 216, and / or the positioning module 270 may be implemented using a combination of hardware and software. In the present disclosure, the operations described as being performed by the DNN 202, the comparison module 216, and / or the positioning module 270 may include operations that the DNN 202, the comparison module 216, and / or the positioning module 270 may instruct the corresponding computing system to perform. In these or other embodiments, DNN 202, comparison module 216, and / or positioning module 270 may be implemented by one or more computing devices, such as Figures 5A-5D , Figure 6 and / or Figure 7 A computing device is described in further detail.

[0066] In some embodiments, the image data 206 may be sent, transmitted, and / or otherwise communicated to the DNN 202. In some embodiments, the image data 206 may be pre-processed, such as by converting, cropping, upscaling, downscaling, zooming, rotating, and / or otherwise modifying the image data 206, before being received by the DNN 202. For example, the image data 206 may be cropped, downscaled, upscaled, flipped, rotated, and / or otherwise adjusted to an input format suitable for the respective DNN 202. In some embodiments, the image data 206 may represent a video (e.g., a video snapshot). For example, any type of image data 206 format may be used, such as a compressed image such as a Joint Photographic Experts Group (JPEG) or Luma / Chro (YUV) format, a compressed image such as a frame derived from a compressed video format such as H.264 / Advanced Video Coding (AVC) or H.265 / High Efficiency Video Coding (HEVC), a raw image such as derived from a red transparent blue (RCCB), red transparent (RCCC), or other type of imaging sensor, and / or other formats. Furthermore, in some examples, image data 206 may be used without any preprocessing (e.g., in a raw or captured format), while in other examples, image data 206 may be preprocessed (e.g., noise balancing, demosaicing, scaling, cropping, enhancement, white balancing, tone curve adjustment, etc.), such as, for example, using one or more preprocessing systems and / or processes.

[0067] In some embodiments, image data 206 may correspond to one or more images that may have been generated using one or more image sensors. In some embodiments, image data 206 may be associated with a device described in the present disclosure, such as, for example, a device for generating an image. Figure 1A The image data 106 further described and / or illustrated is the same and / or similar. Additionally or alternatively, one or more images that may correspond to the image data 206 may be similar to those described in the present disclosure, for example, with respect to Figure 1B Image 120 is further described and / or illustrated.

[0068] In some embodiments, one or more DNNs 202 may be configured to obtain image data 206. In some embodiments, DNNs 202 may be configured to perform one or more operations using image data 206 to determine and / or generate output data 210. Although examples of using DNNs 202 are described herein, this is not intended to be limiting. For example, DNNs 202 may include any type of machine learning model or algorithm, such as a machine learning model using linear regression, logistic regression, decision trees, support vector machines (SVMs), naive Bayes, k-nearest neighbors (Knn), K-means clustering, random forests, dimensionality reduction algorithms, gradient boosting algorithms, neural networks (e.g., autoencoders, convolutions, loops, perceptrons, long / short-term memory / LSTMs, Hopfield, Boltzmann, deep belief, deconvolution, generative adversarial, liquid machines, etc.), region of interest detection algorithms, computer vision algorithms, and / or other types of algorithms or machine learning models.

[0069] For example, the DNN 202 may process the sensor data 102 to generate detections of lane markings, road boundaries, signs, poles, trees, static objects, vehicles, and / or other objects, etc. For example, the detections may correspond to positions (e.g., in 2D image space), geometry, pose, semantic information, and / or other information about the detections.

[0070] In some embodiments, DNN 202 may be configured to determine one or more locations corresponding to representations of one or more objects in image space corresponding to the image. In some embodiments, the one or more locations corresponding to the objects in image space may include one or more points of interest corresponding to the objects. In some embodiments, the points of interest may include reference points corresponding to the objects, and DNN 202 may determine one or more predicted ground locations in image space based on the reference points, which may correspond to 2D map data 208 corresponding to the 2D map.

[0071] In some embodiments, the one or more interest points corresponding to the object may include any portion of the depiction of the object in image space. In some embodiments, for example, the one or more interest points may include a point corresponding to a substantially vertical portion of the object (e.g., a point corresponding to a substantially vertical portion of the object in image space). i In some embodiments, for example, one or more points of interest may correspond to each corner, edge, middle, etc. corresponding to a representation of an object in image space. In some embodiments, one or more points of interest may be determined using one or more heuristic analyses to determine which points of interest produce valid results when comparing one or more predicted image locations to 2D map data 208 corresponding to a 2D map.

[0072] For example, in the context of an autonomous vehicle, DNN 202 may detect one or more road signs corresponding to the environment. Continuing with the example. The one or more points of interest may include one or more locations in image space corresponding to the image. Additionally, the one or more points of interest may include one or more image space locations corresponding to one or more sides of a road sign, a pole supporting the road sign, and any other location corresponding to the road sign.

[0073] Additionally or alternatively, the DNN 202 may be configured to determine one or more predicted image locations associated with one or more determined image locations corresponding to representations of one or more objects. In some embodiments, a reference to one or more predicted image locations may include one or more estimated locations of where objects represented in the 2D map would be in the image space, assuming that a 2D map is superimposed into the image space. The superimposed 2D map represents the object from a perspective corresponding to the estimated pose of the corresponding machine and / or image sensor when the image was captured. For example, one or more predicted image locations may be estimated to coincide with one or more object locations in the image space, wherein the 2D map data 208 may represent an object that is tiled onto a ground plane and projected or superimposed in the image space in a real-world environment.

[0074] In some embodiments, one or more predicted image locations may correspond to one or more points of interest corresponding to objects and ground surfaces in image space that represent a ground surface or ground plane in a real-world environment.

[0075] You can refer to Figure 2B Describe and / or illustrate example implementations of determining one or more predicted image locations corresponding to a particular object, Figure 2B An example diagram showing an image 220, a detected object 212, one or more first predicted image locations 224, and one or more second predicted image locations 226 corresponding to the object 212, according to one or more embodiments of the present disclosure is shown. Although shown in a background corresponding to a road sign on a road, the image 120 may include 2D representations of multiple different environments corresponding to different sensors—for example, referring to Figure 1A The image sensor 102 described herein, and / or one or more machines and / or systems corresponding to different sensors, for example, with reference to Figure 2B Vehicle 118 is described.

[0076] In some embodiments, image 220 may correspond to image data generated and / or collected using one or more sensors. For example, image 220 may correspond to an image data set as described in the present disclosure, such as reference 200. Figure 1A and / or 2A further describes and / or illustrates image data 206 and / or image data 106.

[0077] In some embodiments, image 220 may include origin 214, which may define and / or establish a reference point to which one or more portions of the image space may be related. In these or other embodiments, origin 214 may be related to, for example, Figure 1B The origin 114 described and / or illustrated further is the same and / or similar.

[0078] In some embodiments, image 220 may include one or more objects and / or landmarks, such as object 212. Object 212 in image 220 may be a 2D representation of a sign. Specifically, object 212 may be a speed limit sign corresponding to a road segment. In some embodiments, object 212 may correspond to a 2D representation of a road segment using coordinates (x i ,y i ) defined by the environment. In addition, one or more portions of the object 212 may be identified based on one or more corresponding locations in the image space. For example, the object 212 may include a first side 212a, which may represent a right portion of the object 212. The object 212 may also include a second side 212b, which may represent a left portion of the object 212. In these or other embodiments, the object 212, the first side 212a, and / or the second side 212b may be similar to those described in the present disclosure, for example, with reference to Figure 1B The object 112 , the first side 112 a , and / or the second side 112 b further described and / or illustrated are the same and / or similar.

[0079] In some embodiments, image 220 may include one or more points of interest corresponding to object 212. Figure 2B As shown, the object 212 includes a first interest point that may correspond to a first side 212 a of the object 212 and a second interest point that may correspond to a second side 212 b of the object 212 .

[0080] In some embodiments, image 220 may include horizon 226, which may depict or indicate the difference between a first portion of image 220 (corresponding to the ground in the real-world environment) and a second portion of image 220 (corresponding to a location above the ground in the real-world environment). Figure 2B As shown, a first portion of image 220 (a portion of image 220 that may correspond to the ground in a real-world environment) may include a portion of image 220 that is below horizon 226 .

[0081] In some embodiments, image 220 may not include horizon 226. Rather, image data corresponding to image 220 (e.g., image data 206) may provide one or more other indicators that may distinguish between portions of image 220 that correspond to the ground and portions of the image that may correspond to one or more locations above the ground in a real-world environment.

[0082] In some embodiments, one or more predicted image locations 222 may be determined. In some embodiments, one or more predicted image locations 222 may correspond to one or more points of interest corresponding to object 212.

[0083] In some embodiments, the first predicted image position 222a may correspond to a position located along the y axis from the first side 212a of the object 212. i The axis extends vertically to any point (first point of interest) on a first line segment at the bottom of image 220. In some embodiments, first predicted image positions 222a may correspond only to those positions located on the first line segment bounded by horizon 226 and the bottom of image 220.

[0084] like Figure 2B As shown, the dotted lines extending from the first side 212a and the second side 212b of the object 212 may not be included in the first predicted image position and / or the second predicted image position. Instead, the dotted lines are intended to show that the first predicted image position and the second predicted image position correspond to a line segment corresponding to the ground, and that the interest point corresponding to the object 212 is located on the same line.

[0085] In some embodiments, the first line segment may be bounded only by the horizon 226, and the bottom portion of the image 220 may be extended using other data and / or information corresponding to the environment depicted in the image 220. For example, in the case where the bottom portion of the image 220 is occluded, the object is partially occluded, and / or the ground is occluded, the DNN 202 may determine one or more predicted positions 222 based on other data corresponding to the environment and / or image space (e.g., using past image data corresponding to the environment depicted in the image 220).

[0086] In some embodiments, second predicted image position 222b may correspond to any point located on a second line segment extending vertically from second side 212b of object 212 to a bottom portion of image 220. In some embodiments, second predicted image position 222b may correspond only to those locations located on the second line segment, the second line segment defined by horizon 226 and a bottom portion of image 220. In some embodiments, the second line segment may be defined only by horizon 226, and the bottom portion of image 220 may be extended using other data and / or information corresponding to the environment depicted in image 220.

[0087] In some embodiments, the first predicted image position 222a and / or the second predicted image position 222b may be located at a position that is compatible with y i In some embodiments, the first predicted image position 222a and / or the second predicted image position 222b may be aligned with the y axis. i Axis-aligned or substantially aligned because an image sensor (e.g., a camera) that may have captured image 220 may be oriented perpendicular to a gravity vector corresponding to a real-world environment that may be represented in image 220. There are many variations in which a camera may not be aligned perpendicular to a gravity vector, which variations may be further described in this disclosure, for example with respect to Figure 2A and Figure 2C-2G Describe and / or show.

[0088] Can Figure 2B Modifications, additions, or omissions may be made without departing from the scope of the present disclosure. For example, the placement of origin 214, the environment captured by image 220, the angle of the image sensor capturing image 220, the number of objects 212 and / or the number of parts of objects 212, the placement or existence of horizon 216, etc. may vary. The specifics presented and discussed are intended to aid in explaining and understanding the concepts of the present disclosure and are not intended to be limiting.

[0089] Back to Figure 1A As described in , DNN 202 may be configured to determine one or more predicted image locations based on a vanishing point. The term "vanishing point" may include a point or points in real-world space where parallel lines appear to converge. In some embodiments, one or more equations and / or algorithms may be used to determine one or more predicted image locations based on a projection of a vanishing point in image space.

[0090] For example, the vanishing point may be projected into image space using one or more pinhole projection models to determine where seemingly parallel lines may converge in image space. In some embodiments, for example, the vanishing point may be determined using a gravity vector and one or more intrinsic properties corresponding to the camera (e.g., focal length, calibration determination, etc.). An example equation for projecting the vanishing point into image space is defined as follows:

[0091] (1)

[0092] (2) where v xi represents the projection of the vanishing point in the image space in the x direction, v yi represents the projection of the vanishing point in the image space in the y direction. x , D y and D zRespectively represent the gravity vectors in the X, Y and Z directions corresponding to the real world space. x and f y Respectively represent the focal lengths of the camera lens in the x and y directions in the image space. x and c y Represents a principal point of a camera and / or image sensor that may have generated and / or captured image data 206. In some embodiments, the principal point may include a point corresponding to the center of the optical axis of a camera's lens through which light rays converge to create a focused image on the sensor.

[0093] Additionally, in some embodiments, one or more predicted image locations may be determined based on the ground depiction in image space. In some embodiments, the one or more predicted image locations may correspond to a line segment starting from a point of interest in image space corresponding to the representation of the object and ending at a projected vanishing point in image space. The one or more predicted image locations may correspond to a portion of the line segment that is also located on the ground depiction in image space.

[0094] In some embodiments, Figure 2B For example, in the case where the camera is oriented vertically relative to the gravity vector, the predicted image position may be located at the y coordinate corresponding to the image space. i In the case where the camera and / or image sensor may not be oriented perpendicularly with respect to the gravity vector, for example, where the camera may be rotated along one or more axes, one or more predicted image positions may be located on a line segment associated with one or more points of interest corresponding to a representation of an object in image space and a vanishing point projected into image space.

[0095] You can refer to Figure 2C and Figure 2D Describe and / or illustrate example implementations for determining one or more predicted image locations based on a projected vanishing point corresponding to a particular image, Figure 2C An example diagram representing an image 230 and a vanishing point 228 that may be projected into image space is shown in accordance with one or more embodiments of the present disclosure.

[0096] In some embodiments, image 230 may be an example of one or more images corresponding to image data collected and / or generated using one or more image sensors. In these or other embodiments, image 230 may include representations of one or more points of interest 232 corresponding to one or more objects that may be located in the image space corresponding to image 230.

[0097] Image 230 includes four different points of interest corresponding to one or more objects. First point of interest 232a, second point of interest 232b, third point of interest 232c, and fourth point of interest 232d. As shown in image 230, vanishing point 228 may be projected into the image space corresponding to image 230. In some embodiments, vanishing point 228 may correspond to a point in the image space that is above image 230. In some embodiments, although vanishing point 228 is displayed outside of image 230, the point corresponding to vanishing point 228 is a representation corresponding to a vanishing point in the real world environment projected into the representation of the image space.

[0098] In some embodiments, vanishing point 228 may be located above image 230 in image space due to the orientation of a camera and / or image sensor that may be configured to collect and / or generate image data corresponding to image 230. For example, the camera may be rotated about one or more axes associated with the orientation of the camera in the real-world environment. Thus, vanishing point 228 may be located above image 230 in image space.

[0099] In some embodiments, one or more predicted image locations 234 may be determined by extending a line segment from vanishing point 228 in image space through each of interest points 232 corresponding to image 230. Since one or more predicted image locations 234 correspond to both interest points and a ground representation in image space, one or more predicted image locations 234 may be determined to be located below one or more interest points 232 in image 230, such as Figure 2C and Figure 2D Solid line segments compared to dashed line segments shown in .

[0100] In some embodiments, each point of interest 232 may correspond to one or more predicted image locations 234. Figure 2C As shown, the first interest point 232a may correspond to one or more first predicted image positions 234a, the second interest point 232b may correspond to one or more second predicted image positions 234b, the third interest point 232c may correspond to one or more third predicted image positions 234c, the fourth interest point 232d may correspond to one or more fourth predicted image positions 234d, and so on.

[0101] In some embodiments, the vanishing point may be located below the image. Figure 2DAn example diagram showing image 236 and vanishing point 238 that may be projected below image 236 in image space is shown in accordance with one or more embodiments of the present disclosure. In some embodiments, vanishing point 238 may be located below image 236 in image space due to the orientation of a camera and / or image sensor that may be configured to collect and / or generate image data corresponding to image 236. For example, the camera may be rotated about one or more axes associated with the orientation of the camera in the real-world environment. Thus, vanishing point 238 may be located below image 236 in image space.

[0102] Image 236 may include four different points of interest corresponding to one or more objects: first point of interest 240a, second point of interest 240b, third point of interest 240c, and fourth point of interest 240d. As shown in image 236, vanishing point 238 may be projected into the image space corresponding to image 236.

[0103] In some embodiments, one or more predicted image locations 242 may be determined by extending a line segment in image space from vanishing point 238 to each of interest points 240 corresponding to image 236. Since one or more predicted image locations 242 correspond to both interest points 240 and a ground representation in image space, one or more predicted image locations 242 may be determined to be located below one or more interest points 240 in image 236.

[0104] In some embodiments, each point of interest 240 may correspond to one or more predicted image locations 242. Figure 2D As shown, the first interest point 240a may correspond to one or more first predicted image positions 242a, the second interest point 240b may correspond to one or more second predicted image positions 242b, the third interest point 240c may correspond to one or more third predicted image positions 242c, the fourth interest point 240d may correspond to one or more fourth predicted image positions 242d, and so on.

[0105] Can Figure 2C and / or Figure 2D Modifications, additions, or omissions may be made without departing from the scope of the present disclosure. For example, the location of vanishing point 228 and / or vanishing point 238 in the corresponding image space, the environment captured by image 230 and / or image 236, the angle of the image sensor capturing image 230 and / or image 236, the number of points of interest 232 and / or points of interest 240, the location and / or orientation of one or more predicted image locations 234 and / or one or more predicted image locations 242 may be different. The specifics presented and discussed are intended to aid in explaining and understanding the concepts of the present disclosure and are not intended to be limiting.

[0106] Back to Figure 2A , DNN 202 may also be configured to determine one or more predicted image locations that may correspond to multiple points of interest corresponding to one or more objects in the image. In some embodiments, the number of points of interest may vary. In some embodiments, a single point of interest may be used to determine one or more predicted image locations corresponding to an object. Additionally or alternatively, multiple points of interest may be used to determine one or more predicted image locations corresponding to one or more objects in the image.

[0107] You can refer to Figure 2E , Figure 2F and / or Figure 2G Example implementations of determining one or more points of interest corresponding to a particular object in an image are described and / or illustrated. Figure 2E An example diagram representing an image 244 and one or more predicted image locations 248 corresponding to a vanishing point 246 and a single point of interest associated with an object 250 is shown.

[0108] Figure 2F An example diagram is shown that represents an image 252 and one or more predicted image locations 258 corresponding to a vanishing point 254 and one or more interest points corresponding to an object 256 and the one or more predicted image locations 258. In some embodiments, a first interest point corresponding to the object 256 may correspond to one or more first predicted image locations 258a, a second interest point corresponding to the object 256 may correspond to one or more second predicted image locations 258b, and / or a third interest point corresponding to the object 256 may correspond to one or more third predicted image locations 258c. In some embodiments, the object 256 may correspond to Figure 2E In some embodiments, it may be determined based on one or more heuristic analyses whether using multiple points of interest corresponding to object 250 and / or object 256 may be more efficient in performing one or more positioning determinations. Specifically, whether a single point of interest or multiple points of interest may contribute to one or more cost analyses or comparisons between one or more predicted image locations in the 2D map data corresponding to the 2D map and one or more locations corresponding to the same object. In these or other embodiments, one or more cost analyses and / or comparisons may be performed in the present disclosure, for example, for Figure 1A , Figure 2A or Figure 3 further described and / or shown.

[0109] Figure 2GAn example diagram showing an image 260 and one or more predicted image locations 266 corresponding to a vanishing point 262 and one or more interest points corresponding to an object 264 is shown. Figure 2G In FIG. 2 , object 264 may include the back of a large truck, a road sign, or some other rectangular object 264 .

[0110] In some embodiments, four points of interest may be determined. A first point of interest corresponding to one or more first predicted image locations 266a, a second point of interest corresponding to one or more second predicted image locations 266b, a third point of interest corresponding to one or more third predicted image locations 266c, and a fourth point of interest corresponding to one or more fourth predicted image locations 266d. In some embodiments, the respective corners corresponding to object 264 may correspond to the respective points of interest associated with object 264.

[0111] In some embodiments, one or more heuristic analyses may be performed to determine effective placement and / or determination of points of interest corresponding to objects represented in one or more images. In these or other embodiments, the heuristic analysis may be combined with the methods described in this disclosure, for example, with respect to Figure 2E and Figure 2F The heuristic analysis described and / or illustrated for object 256 and / or object 250 further described is the same and / or similar.

[0112] Can Figure 2E , Figure 2F and / or Figure 2G Modifications, additions, or omissions may be made without departing from the scope of the present disclosure. For example, the environment captured by image 244, image 252, and / or image 260, the location of vanishing point 246, vanishing point 254, and / or vanishing point 262, the number of objects, etc. may be different. The specific content presented and discussed is to help explain and understand the concepts of the present disclosure and is not meant to be limiting.

[0113] Back to Figure 2A , DNN 202 may be configured to generate output data 210 based at least on image data 206. In some embodiments, output data 210 may include data corresponding to one or more images corresponding to image data 206 and / or one or more recognized objects corresponding to one or more images. Additionally or alternatively, output data 210 may include projected vanishing points, points of interest corresponding to one or more objects, and / or one or more predicted image locations.

[0114] In some embodiments, the output data 210 may undergo one or more post-processing operations. In some embodiments, the one or more post-processing operations may include temporal filtering, weighting, outlier removal (e.g., removing pixels or points determined to be outliers), upscaling (e.g., the output may be predicted at a lower resolution than the input sensor data instance, and the output may be upscaled back to the input resolution), downscaling, curve fitting, and / or other post-processing techniques. In embodiments, the output 204 (after post-processing) may be in a 2D coordinate space (e.g., image space).

[0115] In some embodiments, the output data 210 may be transmitted, sent, or otherwise communicated to the comparison module 216. Additionally or alternatively, in some embodiments, the comparison module 216 may be configured to receive and / or otherwise obtain the 2D map data 208.

[0116] In some embodiments, 2D map data 208 may include map data corresponding to a 2D map of an environment in which a machine and / or image sensor associated with positioning system 204 may be located. In some embodiments, 2D map data 208 may also include data corresponding to a pose of the machine and / or image sensor. In these or other embodiments, 2D map data 208 may be associated with, for example, a 2D map of an environment in which a machine and / or image sensor associated with positioning system 204 may be located. Figure 1A The examples of 2D map data 108 described and / or shown further are the same as and / or examples thereof.

[0117] In some embodiments, the comparison module 216 may be configured to compare a 2D visual representation of a map corresponding to the map data 208 with one or more predicted image locations in the image space. For example, a 2D map corresponding to the 2D map data 208 may be projected or otherwise superimposed in the image space corresponding to the image data 206. In some embodiments, projecting a pose corresponding to one or more objects in the 2D map may result in determining a location corresponding to one or more objects in a portion of the image representing the ground in the real-world environment.

[0118] In some embodiments, the comparison module 216 may be configured to determine and / or generate cost data that may indicate a cost or error. In some embodiments, the cost or error may be determined based on a difference between a 2D visual representation of an object and one or more predicted image locations corresponding to the same object in image space corresponding to the image.

[0119] In some embodiments, one or more distance transforms may be used to perform a comparison between one or more predicted image locations in image space and the 2D map data 208 corresponding to the 2D map. For example, one or more distance transform functions may be used to perform a difference between one or more predicted image locations and the 2D map data 208 corresponding to the map. In some embodiments, the distance transform function may include one or more of a Euclidean distance transform, a Manhattan distance transform, a chamfer distance transform, and / or other distance transform operations and / or a function that may be configured to determine a difference between the projected 2D map data 208 corresponding to the 2D map and one or more predicted image locations corresponding to one or more locations of interest associated with the object.

[0120] In some embodiments, the comparison module 216 may be configured to generate cost data 218 based on a comparison between the predicted image position and the 2D map data 208 projected into the image space. In some embodiments, a high cost may indicate that there is a significant difference between the predicted image position and the 2D map data 208 projected into the image space, or a difference that satisfies a threshold. Additionally or alternatively, a low cost or no cost may indicate that the distance between the predicted image position and the projected 2D map data 208 is not significant, or that the distance satisfies a threshold.

[0121] In some embodiments, it may be determined whether and / or to what extent an estimated pose corresponding to a system and / or image sensor used to collect and / or generate image data 206 is incorrect by comparing one or more predicted image positions to 2D map data 208. For example, in some cases, a comparison between one or more predicted image positions and 2D map data 208 corresponding to a 2D map projected into image space may result in high costs or errors, which may indicate that there are significant differences between one or more predicted image positions and 2D map data 208.

[0122] In some embodiments, based on the comparison, it can be determined whether estimated pose data corresponding to the system and / or image sensor in the 2D map data may indicate that the system is too close or too far to one side or the other of one or more objects in image space.

[0123] You can refer to Figure 3 to describe and / or illustrate example implementations of comparing one or more predicted image locations to 2D map data corresponding to a map, Figure 3 An example graph representing a cost comparison between one or more predicted image locations 322 and one or more sets of projected 2D map data corresponding to various points of interest associated with object 312 is shown in accordance with one or more embodiments of the present disclosure.

[0124] In some embodiments, object 312 may be a representation of an object in a real-world environment, represented in image space. In these or other embodiments, object 312 may be similar to any of the embodiments described in this disclosure, for example, with respect to Figure 1B and / or object 112 and / or object 212 further described in 2B are the same and / or similar.

[0125] In some embodiments, image 300 may include origin 314. Origin may be a reference point for other locations in image space corresponding to image 300. In these or other embodiments, origin 314 may be similar to any of the above-described reference points. Figure 1B and Figure 2B The origin 214 and / or origin 114 further described and / or illustrated are the same and / or similar.

[0126] In some embodiments, one or more predicted image locations 322 may correspond to sides of object 312. For example, first side 312a may correspond to one or more first predicted image locations 322a, and second side 312b may correspond to one or more second predicted image locations 322b. In some embodiments, one or more predicted image locations 322 may correspond to the sides of object 312. Figure 2B The predicted image position 222 described and / or shown further is the same and / or similar.

[0127] In some embodiments, image 300 may include map data corresponding to one or more 2D locations of points of interest of object 312 projected into an image space corresponding to image 300. In some embodiments, the one or more 2D locations corresponding to object 312 in the map data may be determined based on an estimated location corresponding to a machine and / or an image sensor in the map data corresponding to the map.

[0128] In some embodiments, the number of projected 2D positions corresponding to object 312 may be associated with the number of points of interest corresponding to object 312. Figure 3 As shown, for example, one or more predicted image locations correspond to two interest points, a first interest point corresponding to a first side 312 a of the object 312 , and a second interest point corresponding to a second side 312 b of the object 312 .

[0129] In some embodiments, first set of projected 2D locations 324 can be compared to one or more predicted image locations 322. In some embodiments, the comparison can determine that there is a difference (in some cases, a significant or costly difference) between first set of projected 2D map locations 324 and one or more predicted image locations 322. In some embodiments, in response to the comparison indicating that the first set of projected 2D map locations are closer than one or more predicted image locations 322, it can be determined that the estimated pose corresponding to the system and / or image sensor may be longitudinally misaligned. For example, it can be determined that the estimated pose of the machine and / or image sensor is too close or too far from object 312 in 2D map data (e.g., 2D map data 208).

[0130] In another example, second set of projected 2D map locations 326 can be compared to one or more predicted image locations 322. In some embodiments, the comparison can determine that there is no difference or a low cost between second set of projected 2D map locations 326 and one or more predicted image locations 322. In some embodiments, the estimated pose of the machine and / or image sensor in the 2D map data is correct compared to image 300.

[0131] In another example instance, third set of projected 2D map locations 328 can be compared to one or more predicted image locations 322. In some embodiments, the comparison can determine that there is a difference (in some instances, a significant or costly difference) between third set of projected 2D map locations 328 and one or more predicted image locations 322. In some embodiments, in response to the comparison indicating that the first set of projected 2D map locations are farther apart than one or more predicted image locations 322, it can be determined that the estimated pose corresponding to the system and / or image sensor may be longitudinally misaligned, just like first set of projected 2D map locations 324. For example, longitudinal misalignment can indicate that the estimated pose of the machine and / or image sensor in the 2D map data is too close or too far from object 312.

[0132] In some embodiments, the fourth set of projected 2D map locations 334 can be compared to the one or more predicted image locations 322. In some embodiments, the comparison can determine that there are differences (in some cases, significant or costly) between the fourth set of projected 2D map locations 334 and the one or more predicted image locations 322. In some embodiments, in response to the comparison indicating that the first set of projected 2D map locations are correctly spaced but aligned to the right of the one or more predicted image locations, it can be determined that the estimated pose corresponding to the system and / or image sensor may be laterally misaligned. For example, the lateral misalignment may indicate that the estimated pose of the machine and / or image sensor in the 2D map data is to the right or left of the object.

[0133] Can Figure 3 Modifications, additions, or omissions may be made without departing from the scope of the present disclosure. For example, the placement of origin 314, the environment captured by image 300, the angle of the image sensor capturing image 300, the number of objects 312 and / or the number of parts of objects 312, the number of projected 2D map locations may vary. The specifics presented and discussed are intended to aid in explaining and understanding the concepts of the present disclosure and are not meant to be limiting.

[0134] Back to Figure 2A , the cost data 218 may be sent, transmitted or otherwise communicated to the positioning module 270. In some embodiments, the positioning module 270 may be configured to determine one or more estimated positions corresponding to the object. In some embodiments, the positioning module 270 may be configured to perform one or more operations using the cost data 218. In some embodiments, the estimated pose corresponding to the system and / or a camera associated with the system may be changed based on the determined cost. In some embodiments, the estimated pose may be iteratively adjusted until the cost meets a specific threshold. In some embodiments, the cost meeting a specific threshold may indicate that the pose corresponding to the system and / or a camera associated with the system in the 2D map data may be correct relative to the object in the environment. In some embodiments, after the specific threshold is met, the positioning module 270 may be configured to generate positioning data 272.

[0135] Positioning data 272 may indicate one or more locations in the 2D map data where the system and / or (one or more) image sensors corresponding to the system may be located. In some embodiments, positioning data 272 may be updated at one or more different timestamps and / or using one or more different images. In some embodiments, positioning data 272 may be transmitted to one or more other systems, subsystems, etc. For example, positioning data 272 may be transmitted to one or more control systems, which may be configured to generate one or more control commands to instruct one or more systems to perform operations, such as, for example, the present disclosure with respect to Figure 1A In some embodiments, the positioning data may be used by one or more computing systems and / or the machine to perform operations such as Figures 5A-5D , Figure 6 and / or Figure 7 The operation is described in further detail.

[0136] Can Figure 2AModifications, additions, or omissions may be made without departing from the scope of the present disclosure. For example, the number of DNN 202 and / or other modules may vary. In addition, the operations described herein may all be performed using one or more of the DNNs and / or modules in positioning system 204. The details given and discussed are intended to aid in explaining and understanding the concepts of the present disclosure and are not meant to be limiting.

[0137] Figure 4 4 is a flow chart illustrating a method 400 for locating one or more systems according to one or more embodiments of the present disclosure. The method 400 may include one or more blocks 402, 404, 406, 408, and 410. Although illustrated as discrete blocks, operations associated with one or more blocks of the method 400 may be divided into additional blocks, combined into fewer blocks, or eliminated, depending on the specific implementation.

[0138] In some embodiments, method 400 may include block 402. At block 402, an image location corresponding to a portion of an object in an image may be determined. In some embodiments, the image location may be defined in a 2D image space. Additionally or alternatively, the 2D image space may be defined based on image data corresponding to the image. In some embodiments, a camera may be used to generate the 2D image space. In some embodiments, the camera used to generate an image corresponding to the 2D image space may be a single monocular camera.

[0139] At block 404, one or more predicted image locations may be determined. In some embodiments, the one or more predicted image locations may correspond to image locations of the representation of the portion of the object. Additionally or alternatively, in some embodiments, the one or more image locations may be determined based on a representation of the ground in the real-world environment, a vanishing point that may be projected in the image space, and the image location of the representation of the portion of the object.

[0140] In some embodiments, the vanishing point may be determined based on the orientation of the camera relative to a gravity vector, where the gravity vector may indicate a direction of gravity corresponding to a position projected into the 2D image space. Furthermore, in some embodiments, one or more characteristics of the camera may be additionally used to determine the vanishing point. In some embodiments, the representation of the one or more predicted image positions may include a line segment that may include at least an image position of a representation of a portion of the object in an image corresponding to the environment and a vanishing point projected into the image space.

[0141] At block 406, one or more ground plane locations of the portion of the object may be compared to the one or more predicted image locations. In some embodiments, the one or more ground plane locations may be determined using a 2D map of the environment that may be projected into the 2D image space. In some embodiments, the one or more ground plane locations may be projected into the 2D image space using a pinhole projection model.

[0142] At block 408, a cost may be determined based on a comparison between the one or more ground plane locations and the one or more predicted image locations. In some embodiments, based on the determined cost, one or more ground plane locations of the portion of the object may be iteratively updated until the determined cost satisfies a cost threshold.

[0143] At block 410 , the system may be positioned to the 2D map data based on the determined cost.

[0144] Without departing from the scope of the present disclosure, the method 400 and / or one or more operations included in the method 400 may be modified, added, or omitted. For example, the operations corresponding to the method 400 may be implemented in a different order. Additionally or alternatively, two or more operations may be performed simultaneously. Furthermore, the operations and actions outlined are provided only as examples, and some of the operations and actions may be optional, combined into fewer operations and actions, or expanded into additional operations and actions without departing from the essence of the described embodiments.

[0145] Example Autonomous Vehicle

[0146] Figure 5Ais an illustration of an example autonomous vehicle 500 according to some embodiments of the present disclosure. Autonomous vehicle 500 (alternatively referred to herein as "vehicle 500") may include, but is not limited to, a passenger vehicle, such as a car, a truck, a bus, an emergency vehicle, a shuttle, an electric or motorized bicycle, a motorcycle, a fire truck, a police car, an ambulance, a boat, an engineering vehicle, a submarine, a drone, and / or other types of vehicles (e.g., unmanned and / or capable of accommodating one or more passengers). Autonomous vehicles are generally described according to the automation levels defined by the National Highway Traffic Safety Administration (NHTSA), a division of the U.S. Department of Transportation, and the Society of Automotive Engineers (SAE) "Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles" (Standard No. J3016-201806 issued on June 15, 2018, Standard No. J3016-201609 issued on September 30, 2016, and previous and future versions of the standard). The vehicle 500 may be capable of implementing one or more functions in accordance with levels 3-5 of autonomous driving levels. The vehicle 500 may be capable of implementing one or more functions in accordance with levels 3-5 of autonomous driving levels. For example, depending on the embodiment, the vehicle 500 may be capable of driver assistance (level 1), semi-automation (level 2), conditional automation (level 3), high automation (level 4) and / or full automation (level 5). The term "autonomy" as used herein may include any and / or all types of autonomy of the vehicle 500 or other machines, such as fully autonomous, highly autonomous, conditionally autonomous, partially autonomous, providing auxiliary autonomy, semi-autonomous, mainly autonomous, or other designations.

[0147] The vehicle 500 may include components such as a chassis, a body, wheels (e.g., 2, 4, 6, 8, 18, etc.), tires, axles, and other components of the vehicle. The vehicle 500 may include a propulsion system 550, such as an internal combustion engine, a hybrid power plant, an all-electric engine, and / or another type of propulsion system. The propulsion system 550 may be connected to a drive train of the vehicle 500, which may include a transmission, to achieve propulsion of the vehicle 500. The propulsion system 550 may be controlled in response to receiving a signal from a throttle / accelerator 552.

[0148] A steering system 554, which may include a steering wheel, may be used to steer vehicle 500 (e.g., along a desired path or route) when propulsion system 550 is operating (e.g., when the vehicle is in motion). Steering system 554 may receive signals from steering actuator 556. For fully automated (Level 5) functionality, a steering wheel may be optional.

[0149] Brake sensor system 546 may be used to operate vehicle brakes in response to receiving signals from brake actuator 548 and / or brake sensors.

[0150] May include one or more CPUs, system on chip (SoC) 504 ( Figure 5C ) and / or one or more controllers 536 of one or more GPUs can provide signals (e.g., representing commands) to one or more components and / or systems of the vehicle 500. For example, the one or more controllers can send signals to operate vehicle brakes via one or more brake actuators 548, to operate the steering system 554 via one or more steering actuators 556, and / or to operate the propulsion system 550 via one or more throttles / accelerators 552. The one or more controllers 536 can include one or more onboard (e.g., integrated) computing devices (e.g., supercomputers) that process sensor signals and output operating commands (e.g., signals representing commands) to enable autonomous driving and / or assist a human driver in driving the vehicle 500. The one or more controllers 536 can include a first controller 536 for autonomous driving functions, a second controller 536 for functional safety functions, a third controller 536 for artificial intelligence functions (e.g., computer vision), a fourth controller 536 for infotainment functions, a fifth controller 536 for redundancy in emergency situations, and / or other controllers. In some examples, a single controller 536 may handle two or more of the above functions, two or more controllers 536 may handle a single function, and / or any combination thereof.

[0151] The one or more controllers 536 may provide signals for controlling one or more components and / or systems of the vehicle 500 in response to sensor data (e.g., sensor input) received from one or more sensors. The sensor data may be received from, for example and without limitation, a global navigation satellite system sensor 558 (e.g., a global positioning system sensor), a RADAR sensor 560, an ultrasonic sensor 562, a LIDAR sensor 564, an inertial measurement unit (IMU) sensor 566 (e.g., an accelerometer, a gyroscope, a magnetic compass, a magnetometer, etc.), a microphone 596, a stereo camera 568, a wide angle camera 570 (e.g., a fisheye camera), an infrared camera 572, a surround camera 574 (e.g., a 360 degree camera), a long-range and / or mid-range camera 598, a speed sensor 544 (e.g., for measuring the velocity of the vehicle 500), a vibration sensor 542, a steering sensor 540, a brake sensor 546 (e.g., as part of a brake sensor system 546), and / or other sensor types.

[0152] One or more of the controllers 536 may receive inputs (e.g., represented by input data) from the instrument cluster 532 of the vehicle 500 and provide outputs (e.g., represented by output data, display data, etc.) via a human machine interface (HMI) display 534, an audible annunciator, a speaker, and / or via other components of the vehicle 500. These outputs may include information such as vehicle speed, velocity, time, map data (e.g., Figure 5C The HMI display 534 may include information such as the HD map 522 of the vehicle 500, location data (e.g., the location of the vehicle 500 on the map), directions, the location of other vehicles (e.g., an occupancy grid), information about objects and object states as sensed by the controller 536, etc. For example, the HMI display 534 may display information about the presence of one or more objects (e.g., street signs, warning signs, traffic light changes, etc.) and / or information about driving maneuvers that the vehicle has made, is making, or will make (e.g., changing lanes now, leaving 34B in two miles, etc.).

[0153] The vehicle 500 further includes a network interface 524 that can communicate via one or more networks using one or more wireless antennas 526 and / or a modem. For example, the network interface 524 may be capable of communicating via LTE, WCDMA, UMTS, GSM, CDMA2000, etc. The one or more wireless antennas 526 may also enable communication between objects in the environment (e.g., vehicles, mobile devices, etc.) using one or more local area networks such as Bluetooth, Bluetooth LE, Z-wave, ZigBee, etc. and / or one or more low power wide area networks (LPWANs) such as LoRaWAN, SigFox, etc.

[0154] Figure 5B For use according to some embodiments of the present disclosure Figure 5A An example of camera positions and fields of view for an example autonomous vehicle 500 is shown. The cameras and respective fields of view are an example embodiment and are not intended to be limiting. For example, additional and / or replaceable cameras may be included, and / or the cameras may be located at different locations on the vehicle 500.

[0155] The camera type for the camera may include, but is not limited to, a digital camera that may be suitable for use with components and / or systems of the vehicle 500. The camera may operate at an automotive safety integrity level (ASIL) B and / or at another ASIL. The camera type may have any image capture rate, such as 60 frames per second (fps), 120fps, 240fps, etc., depending on the embodiment. The camera may be capable of using a rolling shutter, a global shutter, another type of shutter, or a combination thereof. In some examples, the color filter array may include a red-white-white-white (RCCC) color filter array, a red-white-white-blue (RCCB) color filter array, a red-blue-green-white (RBGC) color filter array, a Foveon X3 color filter array, a Bayer sensor (RGGB) color filter array, a monochrome sensor color filter array, and / or another type of color filter array. In some embodiments, a clear pixel camera such as a camera with an RCCC, RCCB, and / or RBGC color filter array may be used in an effort to improve light sensitivity.

[0156] In some examples, one or more of the cameras can be used to perform advanced driver assistance system (ADAS) functions (e.g., as part of a redundant or fail-safe design). For example, a multi-function monocular camera can be installed to provide functions including lane departure warning, traffic sign assistance, and intelligent headlight control. One or more of the cameras (e.g., all of the cameras) can simultaneously record and provide image data (e.g., video).

[0157] One or more of the cameras may be mounted in a mounting assembly such as a custom designed (3-D printed) assembly to cut off stray light and reflections from within the vehicle (e.g., reflections from the dashboard reflected in the windshield mirror) that may interfere with the camera's image data capture capabilities. With respect to the wing mirror mounting assembly, the wing mirror assembly may be custom 3-D printed so that the camera mounting plate matches the shape of the wing mirror. In some examples, one or more cameras may be integrated into the wing mirror. For side view cameras, one or more cameras may also be integrated into the four pillars at each corner of the cab.

[0158] A camera (e.g., a front-facing camera) having a field of view that includes a portion of the environment in front of the vehicle 500 can be used for surround vision to help identify the forward path and obstacles, as well as assist in providing information critical to generating an occupancy grid and / or determining a preferred vehicle path with the help of one or more controllers 536 and / or control SoCs. The front-facing camera can be used to perform many of the same ADAS functions as LIDAR, including emergency braking, pedestrian detection, and collision avoidance. The front-facing camera can also be used for ADAS functions and systems, including lane departure warning ("LDW"), autonomous cruise control ("ACC"), and / or other functions such as traffic sign recognition.

[0159] A variety of cameras may be used in the front-facing configuration, including, for example, a monocular camera platform including a CMOS (complementary metal oxide semiconductor) color imager. Another example may be a wide-angle camera 570, which may be used to sense objects (e.g., pedestrians, intersection traffic, or bicycles) entering the field of view from the periphery. Although Figure 5B Only one wide-angle camera is shown in the figure, but there may be any number of wide-angle cameras 570 on the vehicle 500. In addition, long-range cameras 598 (e.g., a long-view stereo camera pair) may be used for depth-based object detection, especially for objects for which a neural network has not been trained. Long-range cameras 598 may also be used for object detection and classification and basic object tracking.

[0160] One or more stereo cameras 568 may also be included in the front configuration. One or more of the stereo cameras 568 may include an integrated control unit including an expandable processing unit that may provide a multi-core microprocessor and programmable logic ("FPGA") with an integrated CAN or Ethernet interface on a single chip. Such a unit may be used to generate a 3-D map of the vehicle environment, including distance estimates for all points in the image. An alternative stereo camera 568 may include a compact stereo vision sensor that may include two camera lenses (one on each side) and an image processing chip that may measure the distance from the vehicle to the target object and use the generated information (e.g., metadata) to activate autonomous emergency braking and lane departure warning functions. Other types of stereo cameras 568 may be used in addition to or alternatively to those described herein.

[0161] A camera (e.g., a side view camera) having a field of view of a portion of the environment including the sides of the vehicle 500 can be used for surround viewing, providing information used to create and update the occupancy grid and generate side impact collision warnings. Figure 5B The four surround cameras 574 shown in FIG. 5 can be placed on the vehicle 500. The surround cameras 574 can include a wide-angle camera 570, a fisheye camera, a 360-degree camera, and / or the like. For example, the four fisheye cameras can be placed in front, behind, and on the sides of the vehicle. In an alternative arrangement, the vehicle can use three surround cameras 574 (e.g., left, right, and rear), and can utilize one or more other cameras (e.g., a forward-facing camera) as a fourth surround camera.

[0162] A camera having a field of view that includes a portion of the environment behind the vehicle 500 (e.g., a rear view camera) can be used to assist with parking, surround view, rear collision warning, and creating and updating an occupancy grid. A variety of cameras can be used, including but not limited to cameras that are also suitable as front-facing cameras as described herein (e.g., long-range and / or mid-range cameras 598, stereo cameras 568, infrared cameras 572, etc.).

[0163] Figure 5C For use according to some embodiments of the present disclosure Figure 5A Block diagram of an example system architecture of an example autonomous vehicle 500. It should be understood that this arrangement and other arrangements described herein are set forth merely as examples. Other arrangements and elements (e.g., machines, interfaces, functions, sequences, functional groupings, etc.) may be used in addition to or in place of those shown, and some elements may be omitted entirely. Further, many of the elements described herein are functional entities that may be implemented as discrete or distributed components or in combination with other components, and in any appropriate combination and location. The various functions described herein as being performed by an entity may be implemented by hardware, firmware, and / or software. For example, the various functions may be implemented by a processor executing instructions stored in a memory.

[0164] Figure 5C Each of the components, features, and systems of the vehicle 500 is illustrated as being connected via a bus 502. The bus 502 may include a controller area network (CAN) data interface (alternatively, referred to herein as a "CAN bus"). The CAN may be a network inside the vehicle 500 that assists in controlling various features and functions of the vehicle 500, such as the actuation of brakes, acceleration, braking, steering, windshield wipers, and the like. The CAN bus may be configured to have tens or even hundreds of nodes, each with its own unique identifier (e.g., a CAN ID). The CAN bus may be read to find steering wheel angle, ground speed, engine speed per minute (RPM), button position, and / or other vehicle status indicators. The CAN bus may be ASIL B compatible.

[0165] Although bus 502 is described as a CAN bus here, this is not intended to be limiting. For example, in addition to or alternatively to the CAN bus, FlexRay and / or Ethernet can be used. In addition, although bus 502 is represented by a single line, this is not intended to be limiting. For example, there can be any number of buses 502, which may include one or more CAN buses, one or more FlexRay buses, one or more Ethernet buses, and / or one or more other types of buses using different protocols. In some examples, two or more buses 502 can be used to perform different functions, and / or can be used for redundancy. For example, a first bus 502 can be used for a collision avoidance function, and a second bus 502 can be used for drive control. In any example, each bus 502 can communicate with any component of the vehicle 500, and two or more buses 502 can communicate with the same component. In some examples, each SoC 504, each controller 536, and / or each computer in the vehicle can have access to the same input data (e.g., input from a sensor of the vehicle 500), and can be connected to a common bus such as a CAN bus.

[0166] The vehicle 500 may include one or more controllers 536, such as those described herein. Figure 5A Those controllers described. Controller 536 can be used for a variety of functions. Controller 536 can be coupled to any other different components and systems of vehicle 500, and can be used for control of vehicle 500, artificial intelligence of vehicle 500, infotainment for vehicle 500, and / or the like.

[0167] The vehicle 500 may include one or more system on chip (SoC) 504. The SoC 504 may include a CPU 506, a GPU 508, a processor 510, a cache 512, an accelerator 514, a data store 516, and / or other components and features not shown. The SoC 504 may be used to control the vehicle 500 in a variety of platforms and systems. For example, one or more SoCs 504 may be combined with an HD map 522 in a system (e.g., a system of the vehicle 500), and the HD map may be downloaded from one or more servers (e.g., a server) via a network interface 524. Figure 5D one or more servers 578) to obtain map refreshes and / or updates.

[0168] CPU 506 may include a CPU cluster or CPU complex (alternatively, referred to herein as "CCPLEX"). CPU 506 may include multiple cores and / or L2 caches. For example, in some embodiments, CPU 506 may include eight cores in a coherent multiprocessor configuration. In some embodiments, CPU 506 may include four dual-core clusters, each with a dedicated L2 cache (e.g., 2MB L2 cache). CPU 506 (e.g., CCPLEX) may be configured to support simultaneous cluster operations, such that any combination of CPU 506 clusters can be active at any given time.

[0169] CPU 506 may implement power management capabilities including one or more of the following features: each hardware block may be automatically clock gated when idle to save dynamic power; each core clock may be gated when the core is not actively executing instructions due to the execution of WFI / WFE instructions; each core may be independently power gated; each core cluster may be independently clock gated when all cores are clock gated or power gated; and / or each core cluster may be independently power gated when all cores are power gated. CPU 506 may further implement an enhanced algorithm for managing power states, in which allowed power states and expected wake-up times are specified, and hardware / microcode determines the optimal power state to enter for the core, cluster, and CCPLEX. The processing core may support a simplified power state entry sequence in software, with the work being offloaded to the microcode.

[0170] The GPU 508 may include an integrated GPU (alternatively, referred to herein as an "iGPU"). The GPU 508 may be programmable and efficient for parallel workloads. In some examples, the GPU 508 may use an enhanced tensor instruction set. The GPU 508 may include one or more streaming microprocessors, each of which may include an L1 cache (e.g., an L1 cache with at least 96KB storage capacity), and two or more of these streaming microprocessors may share an L2 cache (e.g., an L2 cache with 512KB storage capacity). In some embodiments, the GPU 508 may include at least eight streaming microprocessors. The GPU 508 may use a computing application programming interface (API). In addition, the GPU 508 may use one or more parallel computing platforms and / or programming models (e.g., NVIDIA's CUDA).

[0171] In the case of automotive and embedded use cases, GPU 508 can be power optimized to achieve optimal performance. For example, GPU 508 can be manufactured on fin field effect transistors (FinFETs). However, this is not intended to be limiting, and GPU 508 can be manufactured using other semiconductor manufacturing processes. Each streaming microprocessor can merge several mixed precision processing cores divided into multiple blocks. For example and without limitation, 64 PF32 cores and 32 PF64 cores can be divided into four processing blocks. In such an example, each processing block can be allocated 16 FP32 cores, 8 FP64 cores, 16 INT32 cores, two mixed precision NVIDIA tensor cores for deep learning matrix arithmetic, L0 instruction cache, thread beam (warp) scheduler, dispatch unit and / or 64KB register file. In addition, the streaming microprocessor can include independent parallel integer and floating point data paths to provide efficient execution of workloads using a mix of computation and addressing computations. The streaming microprocessor can include independent thread scheduling capabilities to allow for finer-grained synchronization and collaboration between parallel threads. A streaming microprocessor may include a combined L1 data cache and shared memory unit to increase performance while simplifying programming.

[0172] GPU 508 can include high bandwidth memory (HBM) and / or a 16GB HBM2 memory subsystem that provides a peak memory bandwidth of approximately 900 GB / s in some examples. In some examples, synchronous graphics random access memory (SGRAM), such as fifth generation graphics double data rate synchronous random access memory (GDDR5), can be used in addition to or in lieu of HBM memory.

[0173] The GPU 508 may include unified memory technology that includes access counters to allow memory pages to be more accurately migrated to the processor that accesses them most frequently, thereby improving the efficiency of memory ranges shared between processors. In some examples, address translation service (ATS) support can be used to allow the GPU 508 to directly access the CPU 506 page table. In such an example, when the GPU 508 memory management unit (MMU) experiences a miss, an address translation request can be transmitted to the CPU 506. In response, the CPU 506 can look up the virtual-physical mapping for the address in its page table and transmit the translation back to the GPU 508. In this way, the unified memory technology can allow a single unified virtual address space to be used for the memory of both the CPU 506 and the GPU 508, thereby simplifying GPU 508 programming and porting applications to the GPU 508.

[0174] In addition, GPU 508 may include access counters that can track how often GPU 508 accesses the memory of other processors. The access counters can help ensure that memory pages are moved to the physical memory of the processor that accesses them most frequently.

[0175] SoC 504 may include any number of caches 512, including those described herein. For example, cache 512 may include an L3 cache available to both CPU 506 and GPU 508 (e.g., connected to both CPU 506 and GPU 508). Cache 512 may include a write-back cache that may track the state of a line, for example, by using a cache coherence protocol (e.g., MEI, MESI, MSI, etc.). Depending on the embodiment, the L3 cache may include 4MB or more, although smaller cache sizes may also be used.

[0176] The SoC 504 may include one or more arithmetic logic units (ALUs) that may be used to perform processing for any of a variety of tasks or operations with respect to the vehicle 500, such as processing a DNN. Additionally, the SoC 504 may include a floating point unit (FPU) or other math coprocessor or digital coprocessor type for performing mathematical operations within the system. For example, the SoC 504 may include one or more FPUs integrated as execution units within the CPU 506 and / or GPU 508.

[0177] SoC 504 may include one or more accelerators 514 (e.g., hardware accelerators, software accelerators, or combinations thereof). For example, SoC 504 may include a hardware acceleration cluster, which may include optimized hardware accelerators and / or large on-chip memory. The large on-chip memory (e.g., 4MB SRAM) may enable the hardware acceleration cluster to accelerate neural networks and other calculations. The hardware acceleration cluster may be used to supplement GPU 508 and offload some tasks of GPU 508 (e.g., free up more cycles of GPU 508 for performing other tasks). As an example, accelerator 514 may be used for targeted workloads (e.g., perception, convolutional neural networks (CNNs), etc.) that are sufficiently stable to be easily controlled for acceleration. When used herein, the term "CNN" may include all types of CNNs, including region-based or regional convolutional neural networks (RCNNs) and fast RCNNs (e.g., for object detection).

[0178] The accelerator 514 (e.g., a hardware acceleration cluster) may include a deep learning accelerator (DLA). The DLA may include one or more tensor processing units (TPUs) that may be configured to provide an additional 10 trillion operations per second for deep learning applications and reasoning. The TPU may be an accelerator configured to perform image processing functions (e.g., for CNN, RCNN, etc.) and optimized for performing image processing functions. The DLA may be further optimized for a specific set of neural network types and floating point operations and reasoning. The design of the DLA may provide higher performance per millimeter than a general-purpose GPU, and far exceeds the performance of the CPU. The TPU may perform several functions, including a single instance convolution function, support, for example, INT8, INT16, and FP16 data types for both features and weights, and post-processor functions.

[0179] The DLA can quickly and efficiently execute neural networks, particularly CNNs, on processed or unprocessed data for any of a wide variety of functions, such as, but not limited to: a CNN for object recognition and detection using data from a camera sensor; a CNN for distance estimation using data from a camera sensor; a CNN for emergency vehicle detection and identification and detection using data from a microphone; a CNN for facial recognition and vehicle owner identification using data from a camera sensor; and / or a CNN for safety and / or security-related events.

[0180] The DLA can perform any function of the GPU 508, and by using an inference accelerator, for example, the designer can target any function to either the DLA or the GPU 508. For example, the designer can focus the processing of CNNs and floating point operations on the DLA, and leave other functions to the GPU 508 and / or other accelerators 514.

[0181] The accelerator 514 (e.g., a hardware acceleration cluster) may include a programmable vision accelerator (PVA), which may be alternatively referred to herein as a computer vision accelerator. The PVA may be designed and configured to accelerate computer vision algorithms for advanced driver assistance systems (ADAS), autonomous driving, and / or augmented reality (AR) and / or virtual reality (VR) applications. The PVA may provide a balance between performance and flexibility. For example, each PVA may include, for example and without limitation, any number of reduced instruction set computer (RISC) cores, direct memory access (DMA), and / or any number of vector processors.

[0182] The RISC core can interact with an image sensor (e.g., an image sensor of any camera described herein), an image signal processor, and / or the like. Each of these RISC cores can include any number of memories. Depending on the embodiment, the RISC core can use any of a number of protocols. In some examples, the RISC core can execute a real-time operating system (RTOS). The RISC core can be implemented using one or more integrated circuit devices, application specific integrated circuits (ASICs), and / or storage devices. For example, the RISC core can include an instruction cache and / or a tightly coupled RAM.

[0183] The DMA may enable components of the PVA to access system memory independently of the CPU 506. The DMA may support any number of features used to provide optimizations for the PVA, including, but not limited to, support for multi-dimensional addressing and / or circular addressing. In some examples, the DMA may support addressing in up to six or more dimensions, which may include block width, block height, block depth, horizontal block stepping, vertical block stepping, and / or depth stepping.

[0184] The vector processor can be a programmable processor that can be designed to efficiently and flexibly perform programming for computer vision algorithms and provide signal processing capabilities. In some examples, the PVA can include a PVA core and two vector processing subsystem partitions. The PVA core can include a processor subsystem, one or more DMA engines (e.g., two DMA engines), and / or other peripherals. The vector processing subsystem can operate as the main processing engine of the PVA, and can include a vector processing unit (VPU), an instruction cache, and / or a vector memory (e.g., VMEM). The VPU core can include a digital signal processor, such as, for example, a single instruction multiple data (SIMD), a very long instruction word (VLIW) digital signal processor. The combination of SIMD and VLIW can enhance throughput and rate.

[0185] Each of the vector processors may include an instruction cache and may be coupled to a dedicated memory. As a result, in some examples, each of the vector processors may be configured to execute independently of the other vector processors. In other examples, the vector processors included in a particular PVA may be configured to employ data parallelism. For example, in some embodiments, multiple vector processors included in a single PVA may execute the same computer vision algorithm, but on different regions of an image. In other examples, the vector processors included in a particular PVA may execute different computer vision algorithms simultaneously on the same image, or even execute different algorithms on sequence images or portions of an image. Among other things, any number of PVAs may be included in a hardware acceleration cluster, and any number of vector processors may be included in each of these PVAs. In addition, the PVA may include additional error correction code (ECC) memory to enhance overall system security.

[0186] The accelerator 514 (e.g., a hardware acceleration cluster) may include an on-chip computer vision network and SRAM to provide high bandwidth, low latency SRAM for the accelerator 514. In some examples, the on-chip memory may include at least 4MB of SRAM consisting of, for example and not limited to, eight field-configurable memory blocks that can be accessed by both the PVA and the DLA. Each pair of memory blocks may include an advanced peripheral bus (APB) interface, configuration circuitry, a controller, and a multiplexer. Any type of memory may be used. The PVA and DLA may access the memory via a backbone that provides high-speed memory access to the PVA and DLA. The backbone may include an on-chip computer vision network that interconnects the PVA and DLA to the memory (e.g., using APB).

[0187] The on-chip computer vision network may include an interface that determines that both the PVA and DLA provide ready and valid signals before transmitting any control signals / addresses / data. Such an interface may provide separate phases and separate channels for transmitting control signals / addresses / data, as well as burst communication for continuous data transmission. This type of interface may comply with ISO 26262 or IEC 61508 standards, but other standards and protocols may also be used.

[0188] In some examples, SoC 504 may include a real-time ray tracing hardware accelerator such as described in U.S. Patent Application No. 16 / 101,232 filed on August 10, 2018. The real-time ray tracing hardware accelerator may be used to quickly and efficiently determine the position and range of objects (e.g., within a world model) in order to generate real-time visualization simulations for RADAR signal interpretation, for sound propagation synthesis and / or analysis, for SONAR system simulations, for general wave propagation simulations, for comparison with LIDAR data for the purpose of positioning and / or other functions, and / or for other purposes. In some embodiments, one or more tree traversal units (TTUs) may be used to perform one or more ray tracing related operations.

[0189] The accelerator 514 (e.g., a hardware accelerator cluster) has a wide range of uses in autonomous driving. The PVA can be a programmable visual accelerator that can be used for key processing stages in ADAS and autonomous vehicles. The capabilities of the PVA are a good match for algorithmic domains that require predictable processing, low power, and low latency. In other words, the PVA performs well on semi-dense or dense rule computations, and even on small data sets that require predictable runtimes with low latency and low power. Therefore, in the context of a platform for autonomous vehicles, the PVA is designed to run classic computer vision algorithms because they are efficient at object detection and integer math.

[0190] For example, according to one embodiment of the technology, PVA is used to perform computer stereo vision. In some examples, algorithms based on semi-global matching can be used, but this is not intended to be limiting. Many applications for level 3-5 autonomous driving require instant motion estimation / stereo matching (e.g., structure from motion, pedestrian recognition, lane detection, etc.). PVA can perform computer stereo vision functions on input from two monocular cameras.

[0191] In some examples, PVA can be used to perform dense optical flow. According to the process raw RADAR data (e.g., using 4D fast Fourier transform) to provide processed RADAR. In other examples, PVA is used for time-of-flight depth processing, which is for example by processing raw time-of-flight data to provide processed time-of-flight data.

[0192] DLA can be used to run any type of network to enhance control and driving safety, including, for example, a neural network that outputs a confidence measure for each object detection. Such a confidence value can be interpreted as a probability, or as providing a relative "weight" of each detection compared to other detections. This confidence value enables the system to make further decisions about which detections should be considered true positive detections rather than false positive detections. For example, the system can set a threshold for confidence and only consider detections that exceed the threshold as true positive detections. In an automatic emergency braking (AEB) system, a false positive detection will cause the vehicle to automatically perform emergency braking, which is obviously undesirable. Therefore, only the most confident detection should be considered a trigger for AEB. DLA can run a neural network for regressing confidence values. The neural network can take at least some subset of parameters as its input, such as bounding box dimensions, ground plane estimates obtained (e.g., from another subsystem), inertial measurement unit (IMU) sensor 566 output related to vehicle 500 orientation and distance, 3D position estimates of objects obtained from neural networks and / or other sensors (e.g., LIDAR sensor 564 or RADAR sensor 560), etc.

[0193] SoC 504 may include one or more data stores 516 (e.g., memory). Data store 516 may be on-chip memory of SoC 504 that may store neural networks to be executed on the GPU and / or DLA. In some examples, data store 516 may be large enough to store multiple instances of the neural network for redundancy and safety. Data store 516 may include L2 or L3 cache 512. References to data store 516 may include references to memory associated with PVA, DLA, and / or other accelerators 514 as described herein.

[0194] SoC 504 may include one or more processors 510 (e.g., embedded processors). Processor 510 may include a boot and power management processor, which may be a dedicated processor and subsystem for handling boot power and management functions and related safety implementations. The boot and power management processor may be part of the SoC 504 boot sequence and may provide runtime power management services. The boot power and management processor may provide clock and voltage programming, auxiliary system low power state transitions, SoC 504 thermal and temperature sensor management, and / or SoC 504 power state management. Each temperature sensor may be implemented as a ring oscillator whose output frequency is proportional to the temperature, and SoC 504 may use the ring oscillator to detect the temperature of CPU 506, GPU 508, and / or accelerator 514. If it is determined that the temperature exceeds a threshold, the boot and power management processor may enter a temperature fault routine and place SoC 504 in a lower power state and / or place the vehicle 500 in a driver safety parking mode (e.g., parking the vehicle 500 safely).

[0195] Processor 510 may further include a set of embedded processors that may be used as an audio processing engine. The audio processing engine may be an audio subsystem that allows for full hardware support for multi-channel audio through multiple interfaces and a wide range of flexible audio I / O interfaces. In some examples, the audio processing engine is a dedicated processor core having a digital signal processor with dedicated RAM.

[0196] The processor 510 may further include an always-on processor engine that may provide the necessary hardware features to support low-power sensor management and wake-up use cases. The always-on processor engine may include a processor core, tightly coupled RAM, supporting peripherals (e.g., timers and interrupt controllers), various I / O controller peripherals, and routing logic.

[0197] The processor 510 may further include a safety cluster engine, which includes a dedicated processor subsystem that handles safety management of automotive applications. The safety cluster engine may include two or more processor cores, tightly coupled RAM, supporting peripherals (e.g., timers, interrupt controllers, etc.), and / or routing logic. In safety mode, the two or more cores may operate in lockstep mode and act as a single core with comparison logic to detect any differences between their operations.

[0198] Processor 510 may further include a real-time camera engine, which may include a dedicated processor subsystem for handling real-time camera management.

[0199] Processor 510 may further include a high dynamic range signal processor, which may include an image signal processor, which is a hardware engine that is part of the camera processing pipeline.

[0200] The processor 510 may include a video image compositer, which may be a processing block (e.g., implemented on a microprocessor) that implements video post-processing functions required for the video playback application to produce a final image for the player window. The video image compositer may perform lens distortion correction for the wide-angle camera 570, the surround camera 574, and / or for an in-cab surveillance camera sensor. The in-cab surveillance camera sensor is preferably monitored by a neural network running on another instance of the advanced SoC, configured to recognize in-cab events and respond accordingly. The in-cab system may perform lip reading to activate mobile phone service and place calls, dictate emails, change vehicle destinations, activate or change the vehicle's infotainment system and settings, or provide voice-activated web surfing. Certain functions are only available to the driver when the vehicle is operating in autonomous mode and are disabled in other circumstances.

[0201] The video image compositer may include enhanced temporal noise reduction for both spatial and temporal noise reduction. For example, in the case of motion in the video, the noise reduction appropriately weights the spatial information and reduces the weight of information provided by adjacent frames. In the case where an image or portion of an image does not include motion, the temporal noise reduction performed by the video image compositer may use information from previous images to reduce noise in the current image.

[0202] The video image compositor may also be configured to perform stereoscopic rectification on the input stereoscopic footage frames. The video image compositor may further be used for user interface composition when the operating system desktop is in use and the GPU 508 does not need to continuously render new surfaces. Even when the GPU 508 is powered on and active for 3D rendering, the video image compositor may be used to offload the GPU 508 to improve performance and responsiveness.

[0203] SoC 504 may further include a mobile industry processor interface (MIPI) camera serial interface for receiving video and input from a camera, a high-speed interface, and / or a video input block that may be used for camera and related pixel input functions. SoC 504 may further include an input / output controller that may be controlled by software and may be used to receive I / O signals that are not committed to a specific role.

[0204] SoC 504 may further include a wide range of peripheral interfaces to enable communication with peripherals, audio codecs, power management and / or other devices. SoC 504 may be used to process data from cameras (connected via Gigabit multimedia serial link and Ethernet), sensors (e.g., LIDAR sensor 564, RADAR sensor 560, etc., which may be connected via Ethernet), data from bus 502 (e.g., speed of vehicle 500, steering wheel position, etc.), data from GNSS sensor 558 (connected via Ethernet or CAN bus). SoC 504 may further include dedicated high-performance mass storage controllers, which may include their own DMA engines, and which may be used to free up CPU 506 from routine data management tasks.

[0205] SoC 504 can be an end-to-end platform with a flexible architecture that spans automation levels 3-5, providing a comprehensive functional safety architecture that leverages and efficiently uses computer vision and ADAS technologies to achieve diversity and redundancy, together with deep learning tools to provide a platform for a flexible and reliable driving software stack. SoC 504 can be faster, more reliable, and even more energy efficient and space efficient than conventional systems. For example, when combined with CPU 506, GPU 508, and data storage 516, accelerator 514 can provide a fast and efficient platform for level 3-5 autonomous vehicles.

[0206] The technology thus provides capabilities and functionality that cannot be achieved by conventional systems. For example, computer vision algorithms can be executed on CPUs, which can be configured using high-level programming languages ​​such as the C programming language to execute a wide variety of processing algorithms across a wide variety of visual data. However, CPUs often fail to meet the performance requirements of many computer vision applications, such as those related to, for example, execution time and power consumption. In particular, many CPUs are unable to execute complex object detection algorithms in real time, which is a requirement for in-vehicle ADAS applications and a requirement for practical Level 3-5 autonomous vehicles.

[0207] In contrast to conventional systems, by providing a CPU complex, a GPU complex, and a hardware acceleration cluster, the techniques described herein allow multiple neural networks to be executed simultaneously and / or sequentially, and the results to be combined together to achieve Level 3-5 autonomous driving functions. For example, a CNN executed on a DLA or dGPU (e.g., GPU 520) may include text and word recognition, allowing the supercomputer to read and understand traffic signs, including signs for which a neural network has not been specifically trained. The DLA may further include a neural network capable of recognizing, interpreting, and providing semantic understanding of the signs, and passing that semantic understanding to a path planning module running on the CPU complex.

[0208] As another example, as required for Level 3, 4, or 5 driving, multiple neural networks may be running simultaneously. For example, a warning sign consisting of "Caution: Flashing Lights Indicate Icing Conditions" along with electric lights may be interpreted by several neural networks independently or collectively. The sign itself may be recognized as a traffic sign by a deployed first neural network (e.g., a trained neural network), and the text "Flashing Lights Indicate Icing Conditions" may be interpreted by a deployed second neural network that informs the vehicle's path planning software (preferably executing on a CPU complex) that icing conditions exist when the flashing lights are detected. The flashing lights may be identified by operating a deployed third neural network over multiple frames that informs the vehicle's path planning software of the presence (or absence) of the flashing lights. All three neural networks may be running simultaneously, for example, within the DLA and / or on the GPU 508.

[0209] In some examples, a CNN for facial recognition and owner recognition can use data from the camera sensor to identify the presence of an authorized driver and / or owner of the vehicle 500. The always-on sensor processing engine can be used to unlock the vehicle and turn on the lights when the owner approaches the driver's door, and in security mode, disable the vehicle when the owner leaves the vehicle. In this way, the SoC 504 provides security against theft and / or carjacking.

[0210] In another example, a CNN for emergency vehicle detection and identification can use data from microphone 596 to detect and identify emergency vehicle sirens. In contrast to conventional systems that use general classifiers to detect sirens and manually extract features, SoC 504 uses CNN to classify environmental and urban sounds and to classify visual data. In a preferred embodiment, the CNN running on the DLA is trained to identify the relative closing rate of emergency vehicles (e.g., by using the Doppler effect). The CNN can also be trained to identify emergency vehicles specific to the local area in which the vehicle is operating, as identified by the GNSS sensor 558. Thus, for example, when operating in Europe, the CNN will seek to detect European sirens, and when in the United States, the CNN will seek to identify only North American sirens. Once an emergency vehicle is detected, with the assistance of the ultrasonic sensor 562, the control program can be used to execute emergency vehicle safety routines to slow the vehicle, drive to the side of the road, stop the vehicle, and / or idle the vehicle until the emergency vehicle passes.

[0211] The vehicle may include a CPU 518 (e.g., a discrete CPU or dCPU) that may be coupled to the SoC 504 via a high-speed interconnect (e.g., PCIe). The CPU 518 may include, for example, an X86 processor. The CPU 518 may be used to perform any of a variety of functions, including, for example, arbitrating potentially inconsistent results between ADAS sensors and the SoC 504, and / or monitoring the status and health of the controller 536 and / or the infotainment SoC 530.

[0212] The vehicle 500 may include a GPU 520 (e.g., a discrete GPU or dGPU) that may be coupled to the SoC 504 via a high-speed interconnect (e.g., NVIDIA's NVLINK). The GPU 520 may provide additional artificial intelligence functionality, for example, by executing redundant and / or different neural networks, and may be used to train and / or update the neural network based on inputs (e.g., sensor data) from sensors of the vehicle 500.

[0213] The vehicle 500 may further include a network interface 524, which may include one or more wireless antennas 526 (e.g., one or more wireless antennas for different communication protocols, such as cellular antennas, Bluetooth antennas, etc.). The network interface 524 can be used to enable wireless connections with the cloud (e.g., with a server 578 and / or other network devices), with other vehicles, and / or with computing devices (e.g., a passenger's client device) via the Internet. In order to communicate with other vehicles, a direct link can be established between the two vehicles, and / or an indirect link can be established (e.g., across a network and through the Internet). The direct link can be provided using a vehicle-to-vehicle communication link. The vehicle-to-vehicle communication link can provide the vehicle 500 with information about vehicles approaching the vehicle 500 (e.g., vehicles in front of, to the side of, and / or behind the vehicle 500). This functionality can be part of the collaborative adaptive cruise control functionality of the vehicle 500.

[0214] The network interface 524 may include a SoC that provides modulation and demodulation functions and enables the controller 536 to communicate over a wireless network. The network interface 524 may include a radio frequency front end for up-conversion from baseband to radio frequency and down-conversion from radio frequency to baseband. The frequency conversion may be performed by a known process and / or may be performed using a super-heterodyne process. In some examples, the radio frequency front end function may be provided by a separate chip. The network interface may include a wireless function for communicating via LTE, WCDMA, UMTS, GSM, CDMA2000, Bluetooth, Bluetooth LE, Wi-Fi, Z-wave, ZigBee, LoRaWAN, and / or other wireless protocols.

[0215] The vehicle 500 may further include data storage 528, which may include off-chip storage (e.g., outside the SoC 504). The data storage 528 may include one or more storage elements, including RAM, SRAM, DRAM, VRAM, flash memory, hard disk, and / or other components and / or devices that can store at least one bit of data.

[0216] The vehicle 500 may further include a GNSS sensor 558. The GNSS sensor 558 (e.g., GPS, assisted GPS sensor, differential GPS (DGPS) sensor, etc.) is used to assist in mapping, perception, occupancy grid generation, and / or path planning functions. Any number of GNSS sensors 558 may be used, including, for example and without limitation, GPS using a USB connector with an Ethernet to serial (RS-232) bridge.

[0217] The vehicle 500 may further include a RADAR sensor 560. The RADAR sensor 560 may be used by the vehicle 500 for remote vehicle detection even in darkness and / or inclement weather conditions. The RADAR functional safety level may be ASILB. The RADAR sensor 560 may use CAN and / or bus 502 (e.g., to transmit data generated by the RADAR sensor 560) for control and access to object tracking data, accessing Ethernet in some examples to access raw data. A variety of RADAR sensor types may be used. For example and without limitation, the RADAR sensor 560 may be suitable for front, rear, and side RADAR use. In some examples, a pulse Doppler RADAR sensor is used.

[0218] The RADAR sensor 560 may include different configurations, such as long-range with a narrow field of view, short-range with a wide field of view, short-range side coverage, and the like. In some examples, the long-range RADAR may be used for adaptive cruise control functions. The long-range RADAR system may provide a wide field of view (e.g., within a range of 250m) achieved by two or more independent browsing. The RADAR sensor 560 may help distinguish between static objects and moving objects, and may be used by the ADAS system for emergency braking assistance and forward collision warnings. The long-range RADAR sensor may include a single-station multimode RADAR with multiple (e.g., six or more) fixed RADAR antennas and high-speed CAN and FlexRay interfaces. In an example with six antennas, the central four antennas may create a focused beam pattern designed to record the surroundings of the vehicle 500 at a higher rate with minimal traffic interference from adjacent lanes. The other two antennas may expand the field of view, making it possible to quickly detect vehicles entering or leaving the lane of the vehicle 500.

[0219] As an example, a medium-range RADAR system may include a range of up to 160m (front) or 80m (rear) and a field of view of up to 42 degrees (front) or 150 degrees (rear). A short-range RADAR system may include, but is not limited to, a RADAR sensor designed to be mounted on both ends of a rear bumper. When mounted on both ends of a rear bumper, such a RADAR sensor system may create two beams that continuously monitor the blind spots behind and beside the vehicle.

[0220] Short-range RADAR systems can be used in ADAS systems for blind spot detection and / or lane change assistance.

[0221] The vehicle 500 may further include ultrasonic sensors 562. The ultrasonic sensors 562, which may be placed on the front, rear, and / or sides of the vehicle 500, may be used for parking assistance and / or creating and updating occupancy grids. A variety of ultrasonic sensors 562 may be used, and different ultrasonic sensors 562 may be used for different detection ranges (e.g., 2.5 m, 4 m). The ultrasonic sensors 562 may operate at a functional safety level of ASIL B.

[0222] The vehicle 500 may include a LIDAR sensor 564. The LIDAR sensor 564 may be used for object and pedestrian detection, emergency braking, collision avoidance, and / or other functions. The LIDAR sensor 564 may be ASIL B for functional safety level. In some examples, the vehicle 500 may include multiple LIDAR sensors 564 (e.g., two, four, six, etc.) that may use Ethernet (e.g., to provide data to a Gigabit Ethernet switch).

[0223] In some examples, the LIDAR sensor 564 may be able to provide a list of objects and their distances for a 360-degree field of view. Commercially available LIDAR sensors 564 may have, for example, an advertised range of approximately 500m, an accuracy of 2cm-3cm, and support for 500Mbps Ethernet connections. In some examples, one or more non-protruding LIDAR sensors 564 may be used. In such examples, the LIDAR sensor 564 may be implemented as a small device that can be embedded in the front, back, side, and / or corner of the vehicle 500. In such examples, the LIDAR sensor 564 may provide a field of view of up to 120 degrees horizontally and 35 degrees vertically, with a range of 200m, even for low reflectivity objects. The front-mounted LIDAR sensor 564 may be configured for a horizontal field of view between 45 and 135 degrees.

[0224] In some examples, LIDAR technologies such as 3D flash LIDAR may also be used. 3D flash LIDAR uses flashes of laser as an emission source to illuminate the vehicle's surroundings up to about 200m. The flash LIDAR unit includes a receiver that records the laser pulse transmission time and reflected light on each pixel, which in turn corresponds to the range from the vehicle to the object. Flash LIDAR can allow highly accurate and distortion-free images of the surrounding environment to be generated with each laser flash. In some examples, four flash LIDAR sensors may be deployed, one on each side of the vehicle 500. Available 3D flash LIDAR systems include solid-state 3D staring array LIDAR cameras (e.g., non-browsing LIDAR devices) with no moving parts other than fans. The flash LiDAR device can use 5 nanosecond Class I (eye-safe) laser pulses per frame, and can capture the reflected laser light in the form of a 3D range point cloud and co-registered intensity data. By using flash LIDAR, and because flash LIDAR is a solid-state device with no moving parts, the LIDAR sensor 564 may be less susceptible to motion blur, vibration, and / or shock.

[0225] The vehicle may further include an IMU sensor 566. In some examples, the IMU sensor 566 may be located at the center of the rear axle of the vehicle 500. The IMU sensor 566 may include, for example and without limitation, an accelerometer, a magnetometer, a gyroscope, a magnetic compass, and / or other sensor types. In some examples, for example, in a six-axis application, the IMU sensor 566 may include an accelerometer and a gyroscope, and in a nine-axis application, the IMU sensor 566 may include an accelerometer, a gyroscope, and a magnetometer.

[0226] In some embodiments, the IMU sensor 566 can be implemented as a miniature high-performance GPS-assisted inertial navigation system (GPS / INS) that combines a micro-electromechanical system (MEMS) inertial sensor, a high-sensitivity GPS receiver, and an advanced Kalman filter algorithm to provide estimates of position, velocity, and attitude. Thus, in some examples, the IMU sensor 566 can enable the vehicle 500 to estimate heading by directly observing and correlating velocity changes from the GPS to the IMU sensor 566 without the need for input from a magnetic sensor. In some examples, the IMU sensor 566 and the GNSS sensor 558 can be combined into a single integrated unit.

[0227] The vehicle may include microphones 596 positioned in and / or around the vehicle 500. The microphones 596 may be used for, among other things, emergency vehicle detection and identification.

[0228] The vehicle may further include any number of camera types, including a stereo camera 568, a wide angle camera 570, an infrared camera 572, a surround camera 574, a long-range and / or mid-range camera 598, and / or other camera types. These cameras may be used to capture image data around the entire periphery of the vehicle 500. The type of camera used depends on the embodiment and the requirements of the vehicle 500, and any combination of camera types may be used to provide the necessary coverage around the vehicle 500. In addition, the number of cameras may vary depending on the embodiment. For example, the vehicle may include six cameras, seven cameras, ten cameras, twelve cameras, and / or another number of cameras. As an example and not limitation, the cameras may support Gigabit Multimedia Serial Link (GMSL) and / or Gigabit Ethernet. Each of the cameras described herein may be described in detail with respect to Figure 5A and Figure 5B Described in more detail.

[0229] The vehicle 500 may further include a vibration sensor 542. The vibration sensor 542 may measure vibrations of components of the vehicle, such as an axle. For example, changes in vibration may indicate changes in the road surface. In another example, when two or more vibration sensors 542 are used, the difference between the vibrations may be used to determine friction or slip of the road surface (e.g., when there is a vibration difference between a powered drive shaft and a free-spinning shaft).

[0230] The vehicle 500 may include an ADAS system 538. In some examples, the ADAS system 538 may include a SoC. The ADAS system 538 may include autonomous / adaptive / automatic cruise control (ACC), cooperative adaptive cruise control (CACC), forward collision warning (FCW), automatic emergency braking (AEB), lane departure warning (LDW), lane keeping assist (LKA), blind spot warning (BSW), rear cross traffic warning (RCTW), collision warning system (CWS), lane centering (LC), and / or other features and functions.

[0231] The ACC system may use a RADAR sensor 560, a LIDAR sensor 564, and / or a camera. The ACC system may include a longitudinal ACC and / or a lateral ACC. The longitudinal ACC monitors and controls the distance to the vehicle immediately in front of the vehicle 500, and automatically adjusts the vehicle speed to maintain a safe distance from the vehicle in front. The lateral ACC performs distance keeping and suggests that the vehicle 500 change lanes when necessary. The lateral ACC is related to other ADAS applications such as LCA and CWS.

[0232] CACC uses information from other vehicles, which can be received from other vehicles indirectly via a wireless link or through a network connection (e.g., through the Internet) via a network interface 524 and / or a wireless antenna 526. A direct link can be provided by a vehicle-to-vehicle (V2V) communication link, while an indirect link can be an infrastructure-to-vehicle (I2V) communication link. Typically, the V2V communication concept provides information about the vehicle immediately ahead (e.g., the vehicle immediately ahead of the vehicle 500 and in the same lane as it), while the I2V communication concept provides information about traffic farther ahead. The CACC system may include either or both of the I2V and V2V information sources. Given information about the vehicle ahead of the vehicle 500, CACC can be more reliable, and it is possible to improve the smoothness of traffic flow and reduce road congestion.

[0233] The FCW system is designed to alert the driver to hazards so that the driver can take corrective action. The FCW system uses a front camera and / or RADAR sensor 560 coupled to a dedicated processor, DSP, FPGA and / or ASIC, which is electrically coupled to driver feedback such as a display, speaker and / or vibrating components. The FCW system can provide warnings in the form of, for example, sound, visual warnings, vibrations and / or rapid brake pulses.

[0234] The AEB system detects an impending front collision with another vehicle or other object and can automatically apply the brakes if the driver does not take corrective action within specified time or distance parameters. The AEB system can use a front camera and / or RADAR sensor 560 coupled to a dedicated processor, DSP, FPGA and / or ASIC. When the AEB system detects a hazard, it typically first alerts the driver to take corrective action to avoid the collision, and if the driver does not take corrective action, the AEB system can automatically apply the brakes in an effort to prevent or at least mitigate the effects of the predicted collision. The AEB system may include technologies such as dynamic brake support and / or collision approach braking.

[0235] The LDW system provides visual, audible, and / or tactile warnings, such as steering wheel or seat vibrations, to alert the driver when the vehicle 500 crosses a lane marking. When the driver indicates an intention to leave the lane, by activating a turn signal, the LDW system is not activated. The LDW system may use a front-facing camera coupled to a dedicated processor, DSP, FPGA, and / or ASIC that is electrically coupled to driver feedback such as a display, speaker, and / or vibrating component.

[0236] The LKA system is a variation of the LDW system. If the vehicle 500 begins to leave the lane, the LKA system provides steering input or braking to correct the vehicle 500. The BSW system detects and warns the driver of vehicles in the car's blind spot. The BSW system can provide visual, audible and / or tactile alerts to indicate that it is unsafe to merge or change lanes. The system can provide additional warnings when the driver uses a turn signal. The BSW system can use a rear-facing camera and / or RADAR sensor.

[0237] The RCTW system can provide visual, audible, and / or tactile notifications when an object is detected outside the range of the rear camera while the vehicle 500 is in reverse. Some RCTW systems include AEB to ensure that the vehicle brakes are applied to avoid a crash. The RCTW system can use one or more rear RADAR sensors 560 coupled to a dedicated processor, DSP, FPGA, and / or ASIC that is electrically coupled to driver feedback such as a display, speaker, and / or vibrating component.

[0238] Conventional ADAS systems may be prone to false positive results, which may annoy and distract the driver, but are typically not catastrophic because the ADAS system alerts the driver and allows the driver to decide whether the safety condition actually exists and take action accordingly. However, in the autonomous vehicle 500, in the case of conflicting results, the vehicle 500 itself must decide whether to pay attention to the results from the main computer or the auxiliary computer (e.g., the first controller 536 or the second controller 536). For example, in some embodiments, the ADAS system 538 can be a backup and / or auxiliary computer for providing perception information to the backup computer rationality module. The backup computer rationality monitor can run redundant and diverse software on hardware components to detect faults in perception and dynamic driving tasks. The output from the ADAS system 538 can be provided to the supervisory MCU. If the outputs from the main computer and the auxiliary computer conflict, the supervisory MCU must determine how to coordinate the conflict to ensure safe operation.

[0239] In some examples, the master computer can be configured to provide a confidence score to the supervisory MCU, indicating the master computer's confidence in the selected result. If the confidence score exceeds a threshold, the supervisory MCU can follow the direction of the master computer, regardless of whether the auxiliary computer provides conflicting or inconsistent results. In the case where the confidence score does not meet the threshold and the master computer and the auxiliary computer indicate different results (e.g., conflicts), the supervisory MCU can arbitrate between these computers to determine the appropriate result.

[0240] The supervisory MCU may be configured to run a neural network that is trained and configured to determine conditions under which the auxiliary computer provides a false alarm based on outputs from the primary computer and the auxiliary computer. Thus, the neural network in the supervisory MCU may learn when the output of the auxiliary computer may be trusted and when it may not. For example, when the auxiliary computer is a RADAR-based FCW system, the neural network in the supervisory MCU may learn when the FCW system is identifying a metal object that is not actually dangerous, such as a drainage grate or manhole cover that triggers an alarm. Similarly, when the auxiliary computer is a camera-based LDW system, the neural network in the supervisory MCU may learn to ignore the LDW when a cyclist or pedestrian is present and lane departure is actually the safest strategy. In an embodiment that includes a neural network running on the supervisory MCU, the supervisory MCU may include at least one of a DLA or a GPU suitable for running the neural network with associated memory. In a preferred embodiment, the supervisory MCU may include and / or be included as a component of the SoC 504.

[0241] In other examples, the ADAS system 538 may include an auxiliary computer that uses traditional computer vision rules to perform ADAS functions. In this way, the auxiliary computer can use classic computer vision rules (if-then), and the presence of a neural network in the supervisory MCU can improve reliability, safety, and performance. For example, diverse implementations and intentional non-identity make the entire system more fault-tolerant, especially for failures caused by software (or software-hardware interface) functions. For example, if there is a software vulnerability or error in the software running on the main computer and the non-identical software code running on the auxiliary computer provides the same overall result, the supervisory MCU can be more confident that the overall result is correct and that the vulnerability in the software or hardware on the main computer does not cause a substantial error.

[0242] In some examples, the output of the ADAS system 538 can be fed to the perception block of the main computer and / or the dynamic driving task block of the main computer. For example, if the ADAS system 538 indicates a forward collision warning due to an object immediately ahead, the perception block can use this information in identifying the object. In other examples, the auxiliary computer can have its own neural network that is trained and thus reduces the risk of false positives as described herein.

[0243] The vehicle 500 may further include an infotainment SoC 530 (e.g., an in-vehicle infotainment system (IVI)). Although illustrated and described as an SoC, the infotainment system may not be an SoC and may include two or more discrete components. The infotainment SoC 530 may include a combination of hardware and software that may be used to provide audio (e.g., music, personal digital assistant, navigation instructions, news, radio, etc.), video (e.g., TV, movies, streaming, etc.), phone (e.g., hands-free calling), network connectivity (e.g., LTE, Wi-Fi, etc.), and / or information services (e.g., navigation system, rear parking assistance, radio data system, vehicle-related information such as fuel level, total distance covered, brake fuel level, oil level, door open / closed, air filter information, etc.) to the vehicle 500. For example, the infotainment SoC 530 may include a radio, a disc player, a navigation system, a video player, USB and Bluetooth connectivity, an onboard computer, onboard entertainment, Wi-Fi, steering wheel audio controls, hands-free voice controls, a head-up display (HUD), an HMI display 534, a telematics device, a control panel (e.g., for controlling and / or interacting with various components, features, and / or systems), and / or other components. The infotainment SoC 530 may further be used to provide information (e.g., visual and / or auditory) to a user of the vehicle, such as information from an ADAS system 538, autonomous driving information such as planned vehicle maneuvers, trajectories, surrounding environment information (e.g., intersection information, vehicle information, road information, etc.), and / or other information.

[0244] The infotainment SoC 530 may include GPU functionality. The infotainment SoC 530 may communicate with other devices, systems, and / or components of the vehicle 500 via a bus 502 (e.g., a CAN bus, Ethernet, etc.). In some examples, the infotainment SoC 530 may be coupled to a supervisory MCU so that in the event of a failure of a main controller 536 (e.g., a main and / or backup computer of the vehicle 500), the GPU of the infotainment system may perform some self-driving functions. In such an example, the infotainment SoC 530 may place the vehicle 500 in a driver safety parking mode as described herein.

[0245] The vehicle 500 may further include an instrument cluster 532 (e.g., a digital instrument panel, an electronic instrument cluster, a digital instrument panel, etc.). The instrument cluster 532 may include a controller and / or a supercomputer (e.g., a separate controller or a supercomputer). The instrument cluster 532 may include a set of instruments, such as a speedometer, a fuel level, an oil pressure, a tachometer, an odometer, a turn indicator, a shift position indicator, a seat belt warning light, a parking brake warning light, an engine fault light, an airbag (SRS) system information, lighting controls, safety system controls, navigation information, and the like. In some examples, information may be displayed and / or shared between the infotainment SoC 530 and the instrument cluster 532. In other words, the instrument cluster 532 may be included as part of the infotainment SoC 530, or vice versa.

[0246] Figure 5D For a cloud-based server and Figure 5A 500. System 576 may include a server 578, a network 590, and a vehicle including the vehicle 500. The server 578 may include a plurality of GPUs 584(A)-584(H) (collectively referred to herein as GPUs 584), PCIe switches 582(A)-582(H) (collectively referred to herein as PCIe switches 582), and / or CPUs 580(A)-580(B) (collectively referred to herein as CPUs 580). The GPUs 584, CPUs 580, and PCIe switches may be interconnected with a high-speed interconnect and / or PCIe connection 586 such as, for example and without limitation, an NVLink interface 588 developed by NVIDIA. In some examples, the GPUs 584 are connected via NVLink and / or NVSwitch SoCs, and the GPUs 584 and PCIe switches 582 are connected via PCIe interconnects. Although eight GPUs 584, two CPUs 580, and two PCIe switches are illustrated, this is not intended to be limiting. Depending on the embodiment, each of the servers 578 may include any number of GPUs 584, CPUs 580, and / or PCIe switches. For example, each of the servers 578 may include eight, sixteen, thirty-two, and / or more GPUs 584.

[0247] Server 578 may receive image data over network 590 and from a vehicle, the image data representing images showing unexpected or changed road conditions, such as recently begun road work. Server 578 may transmit neural network 592, updated neural network 592, and / or map information 594, including information about traffic and road conditions, over network 590 and to the vehicle. Updates to map information 594 may include updates to HD map 522, such as information about construction sites, potholes, curves, flooding, or other obstacles. In some examples, neural network 592, updated neural network 592, and / or map information 594 may have been generated from new training and / or data received from any number of vehicles in the environment and / or based on experience from training performed at a data center (e.g., using server 578 and / or other servers).

[0248] Server 578 can be used to train a machine learning model (e.g., a neural network) based on training data. The training data can be generated by the vehicle, and / or can be generated in a simulation (e.g., using a game engine). In some examples, the training data is labeled (e.g., when the neural network benefits from supervised learning) and / or undergoes other preprocessing, while in other examples, the training data is not labeled and / or preprocessed (e.g., when the neural network does not require supervised learning). Training can be performed according to any one or more classes of machine learning techniques, including but not limited to the following classes: supervised training, semi-supervised training, unsupervised training, self-learning, reinforcement learning, joint learning, transfer learning, feature learning (including principal component and cluster analysis), multilinear subspace learning, manifold learning, representation learning (including alternate dictionary learning), rule-based machine learning, anomaly detection, and any variant or combination thereof. Once the machine learning model is trained, the machine learning model can be used by the vehicle (e.g., transmitted to the vehicle via network 590), and / or the machine learning model can be used by server 578 to remotely monitor the vehicle.

[0249] In some examples, server 578 can receive data from the vehicle and apply the data to the latest real-time neural network for real-time intelligent reasoning. Server 578 may include a deep learning supercomputer and / or a dedicated AI computer powered by GPU 584, such as DGX and DGX Station machines developed by NVIDIA. However, in some examples, server 578 may include a deep learning infrastructure of a data center powered only by CPUs.

[0250] The deep learning infrastructure of server 578 may be capable of rapid real-time inference, and may use this capability to assess and verify the health of the processors, software, and / or associated hardware in vehicle 500. For example, the deep learning infrastructure may receive periodic updates from vehicle 500, such as a sequence of images and / or objects located in the sequence of images that vehicle 500 has located (e.g., via computer vision and / or other machine learning object classification techniques). The deep learning infrastructure may run its own neural network to identify objects and compare them to the objects identified by vehicle 500, and if the results do not match and the infrastructure concludes that the AI ​​in vehicle 500 has failed, then server 578 may transmit a signal to vehicle 500 instructing the fail-safe computer of vehicle 500 to take control, notify passengers, and complete a safe parking maneuver.

[0251] For reasoning, the server 578 may include a GPU 584 and one or more programmable reasoning accelerators (e.g., NVIDIA's TensorRT 3). The combination of GPU-powered servers and reasoning acceleration can make real-time responses possible. In other examples, such as when performance is not so important, CPU, FPGA, and other processor-powered servers can be used for reasoning.

[0252] Example computing device

[0253] Figure 6 6 is a block diagram of an example computing device 600 suitable for implementing some embodiments of the present disclosure. The computing device 600 may include an interconnect system 602 that directly or indirectly couples the following devices: a memory 604, one or more central processing units (CPUs) 606, one or more graphics processing units (GPUs) 608, a communication interface 610, an input / output (I / O) port 612, an input / output component 614, a power supply 616, one or more presentation components 618 (e.g., a display), and one or more logic units 620. In at least one embodiment, the computing device 600 may include one or more virtual machines (VMs), and / or any of its components may include virtual components (e.g., virtual hardware components). For non-limiting examples, the one or more GPUs 608 may include one or more vGPUs, the one or more CPUs 606 may include one or more vCPUs, and / or the one or more logic units 620 may include one or more virtual logic units. Thus, computing device 600 may include discrete components (eg, a complete GPU dedicated to computing device 600 ), virtual components (eg, a portion of a GPU dedicated to computing device 600 ), or a combination thereof.

[0254] although Figure 6The various blocks of are shown as being connected via an interconnect system 602 having wires, but this is not intended to be limiting and is merely for clarity. For example, in some embodiments, a presentation component 618 such as a display device may be considered an I / O component 614 (e.g., if the display is a touch screen). As another example, CPU 606 and / or GPU 608 may include memory (e.g., memory 604 may represent a storage device in addition to the memory of GPU 608, CPU 606, and / or other components). In other words, Figure 6 The computing devices described herein are illustrative only. No distinction is made between categories such as "workstations," "servers," "laptops," "desktops," "tablets," "client devices," "mobile devices," "handheld devices," "game consoles," "electronic control units (ECUs)," "virtual reality systems," and / or other device or system types, as all are considered Figure 6 within the range of computing devices.

[0255] The interconnection system 602 may represent one or more links or buses, such as an address bus, a data bus, a control bus, or a combination thereof. The interconnection system 602 may include one or more links or bus types, such as an industry standard architecture (ISA) bus, an extended industry standard architecture (EISA) bus, a video electronics standard association (VESA) bus, a peripheral component interconnect (PCI) bus, a peripheral component interconnect express (PCIe) bus, and / or another type of bus or link. In some embodiments, there is a direct connection between components. As an example, the CPU 606 may be directly connected to the memory 604. In addition, the CPU 606 may be directly connected to the GPU 608. In the case where there is a direct or point-to-point connection between components, the interconnection system 602 may include a PCIe link to perform the connection. In these examples, the PCI bus does not need to be included in the computing device 600.

[0256] Memory 604 may include any of a variety of computer-readable media. Computer-readable media may be any available media that can be accessed by computing device 600. Computer-readable media may include volatile and nonvolatile media and removable and non-removable media. By way of example and not limitation, computer-readable media may include computer storage media and communication media.

[0257] Computer storage media may include volatile and nonvolatile media and / or removable and non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, program modules, and / or other data types. For example, memory 604 may store computer-readable instructions (e.g., representing programs and / or program elements, such as an operating system). Computer storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other storage technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by computing device 600. As used herein, computer storage media does not include the signals themselves.

[0258] Computer storage media may include computer readable instructions, data structures, program modules, and / or other data types in a modulated data signal such as a carrier wave or other transmission mechanism, and include any information delivery media. The term "modulated data signal" may refer to a signal that has one or more of its characteristics set or changed in such a way that information is encoded into the signal. By way of example and not limitation, computer storage media may include wired media such as a wired network or a direct wired connection, and wireless media such as sound, RF, infrared, and other wireless media. Any combination of the above should also be included within the scope of computer readable media.

[0259] The CPU 606 may be configured to execute at least some of the computer-readable instructions so as to control one or more components of the computing device 600 to perform one or more of the methods and / or processes described herein. Each of the CPUs 606 may include one or more cores (e.g., one, two, four, eight, twenty-eight, seventy-two, etc.) capable of simultaneously processing a large number of software threads. The CPU 606 may include any type of processor, and may include different types of processors, depending on the type of computing device 600 implemented (e.g., a processor with fewer cores for mobile devices and a processor with more cores for servers). For example, depending on the type of computing device 600, the processor may be an Advanced RISC Mechanism (ARM) processor implemented using Reduced Instruction Set Computing (RISC) or an x86 processor implemented using Complex Instruction Set Computing (CISC). The computing device 600 may also include one or more CPUs 606 in addition to one or more microprocessors or supplementary coprocessors such as math coprocessors.

[0260] In addition to or in place of the CPU 606, the GPU 608 may also be configured to execute at least some computer-readable instructions to control one or more components of the computing device 600 to perform one or more methods and / or processes described herein. One or more GPUs 608 may be integrated GPUs (e.g., with one or more CPUs 606) and / or one or more GPUs 608 may be discrete GPUs. In an embodiment, one or more GPUs 608 may be coprocessors of one or more CPUs 606. The computing device 600 may use the GPU 608 to render graphics (e.g., 3D graphics) or perform general-purpose computing. For example, the GPU 608 may be used for general-purpose computing on the GPU (GPGPU). The GPU 608 may include hundreds or thousands of cores capable of processing hundreds or thousands of software threads simultaneously. The GPU 608 may generate pixel data for outputting an image in response to a rendering command (e.g., a rendering command from the CPU 606 received via a host interface). The GPU 608 may include a graphics memory such as a display memory for storing pixel data or any other suitable data (e.g., GPGPU data). Display memory may be included as part of memory 604. GPU 608 may include two or more GPUs operating in parallel (e.g., via a link). The link may connect the GPUs directly (e.g., using NVLINK) or through a switch (e.g., using NVSwitch). When combined together, each GPU 608 may generate pixel data or GPGPU data for different portions of output or for different outputs (e.g., a first GPU for a first image and a second GPU for a second image). Each GPU may include its own memory or may share memory with other GPUs.

[0261] In addition to or in lieu of the CPU 606 and / or GPU 608, the logic unit 620 may be configured to execute at least some computer-readable instructions to control one or more components of the computing device 600 to perform one or more methods and / or processes described herein. In an embodiment, the CPU 606, the GPU 608, and / or the logic unit 620 may discretely or jointly perform any combination of methods, processes, and / or portions thereof. The one or more logic units 620 may be a part of and / or integrated into the one or more CPUs 606 and / or the one or more GPUs 608, and / or the one or more logic units 620 may be discrete components of the CPU 606 and / or the GPU 608 or otherwise external thereto. In an embodiment, the one or more logic units 620 may be a processor of the one or more CPUs 606 and / or the one or more GPUs 608.

[0262] Examples of logic unit 620 include one or more processing cores and / or components thereof, such as a data processing unit (DPU), a tensor core (TC), a tensor processing unit (TPU), a pixel vision core (PVC), a vision processing unit (VPU), a graphics processing cluster (GPC), a texture processing cluster (TPC), a streaming multiprocessor (SM), a tree traversal unit (TTU), an artificial intelligence accelerator (AIA), a deep learning accelerator (DLA), an arithmetic logic unit (ALU), an application specific integrated circuit (ASIC), a floating point unit (FPU), an input / output (I / O) element, a peripheral component interconnect (PCI) or a peripheral component interconnect express (PCIe) element, etc.

[0263] The communication interface 610 may include one or more receivers, transmitters, and / or transceivers that enable the computing device 600 to communicate with other computing devices via an electronic communication network, including wired and / or wireless communications. The communication interface 610 may include components and functions that enable communication over any of a number of different networks, such as wireless networks (e.g., Wi-Fi, Z-wave, Bluetooth, Bluetooth LE, ZigBee, etc.), wired networks (e.g., communicating over Ethernet or InfiniBand), low power wide area networks (e.g., LoRaWAN, SigFox, etc.), and / or the Internet. In one or more embodiments, the one or more logic units 620 and / or the communication interface 610 may include one or more data processing units (DPUs) for transmitting data received over the network and / or interconnect system 602 directly to one or more GPUs 608 (e.g., memory of GPUs 608).

[0264] The I / O port 612 can enable the computing device 600 to be logically coupled to other devices including I / O components 614, presentation components 618, and / or other components, some of which can be built into (e.g., integrated into) the computing device 600. Illustrative I / O components 614 include microphones, mice, keyboards, joysticks, game pads, game controllers, satellite dishes, browsers, printers, wireless devices, and the like. The I / O components 614 can provide a natural user interface (NUI) that processes air gestures, voice, or other physiological inputs generated by a user. In some instances, the input can be transmitted to an appropriate network element for further processing. The NUI can implement any combination of voice recognition, stylus recognition, facial recognition, biometric recognition, gesture recognition on and adjacent to the screen, air gestures, head and eye tracking, and touch recognition associated with the display of the computing device 600 (as described in more detail in this disclosure). The computing device 600 may include a depth camera such as a stereo camera system, an infrared camera system, an RGB camera system, touch screen technology, and a combination of these for gesture detection and recognition. Additionally, computing device 600 may include an accelerometer or gyroscope to enable motion detection (e.g., as part of an inertial measurement unit (IMU)). In some examples, the output of the accelerometer or gyroscope may be used by computing device 600 to render immersive augmented reality or virtual reality.

[0265] The power supply 616 may include a hardwired power supply, a battery power supply, or a combination thereof. The power supply 616 may provide power to the computing device 600 to enable the components of the computing device 600 to operate.

[0266] The presentation component 618 may include a display (e.g., a monitor, a touch screen, a television screen, a head-up display (HUD), other display types, or a combination thereof), a speaker, and / or other presentation components. The presentation component 618 may receive data from other components (e.g., GPU 608, CPU 606, etc.) and output the data (e.g., as an image, video, sound, etc.).

[0267] Sample Data Center

[0268] Figure 7 An example data center 700 is shown, which may be used in at least one embodiment of the present disclosure. The data center 700 may include a data center infrastructure layer 710, a framework layer 720, a software layer 730, and an application layer 740.

[0269] like Figure 7As shown, the data center infrastructure layer 710 may include a resource coordinator 712, grouped computing resources 714, and node computing resources ("node CRs") 716(1)-716(N), where "N" represents any complete positive integer. In at least one embodiment, the node CRs 716(1)-716(N) may include, but are not limited to, any number of central processing units ("CPUs") or other processors (including DPUs, accelerators, field programmable gate arrays (FPGAs), graphics processors or graphics processing units (GPUs), etc.), memory devices (e.g., dynamic read-only memories), storage devices (e.g., solid-state drives or disk drives), network input / output ("NW I / O") devices, network switches, virtual machines ("VMs"), power modules and cooling modules, etc. In some embodiments, one or more of the node CRs 716(1)-716(N) may correspond to a server having one or more of the above computing resources. Furthermore, in some embodiments, nodes CR 716 ( 1 )- 716 (N) may include one or more virtual components, such as vGPU, vCPU, etc., and / or one or more of nodes CR 716 ( 1 )- 716 (N) may correspond to a virtual machine (VM).

[0270] In at least one embodiment, the computing resources 714 of the grouping can include a separate grouping (not shown) of the node CR716 contained in one or more racks, or a lot of racks (not shown) contained in the data center of each geographical location. The separate grouping of the node CR716 in the computing resources 714 of the grouping can include the computing, network, memory or storage resources that can be configured or assigned to support the grouping of one or more workloads. In at least one embodiment, several node CR716 including CPU, GPU, DPU and / or other processors can be grouped in one or more racks to provide computing resources to support one or more workloads. One or more racks can also include any number of power modules, cooling modules and / or network switches in any combination.

[0271] Resource coordinator 712 may configure or otherwise control one or more nodes CR 716(1)-716(N) and / or grouped computing resources 714. In at least one embodiment, resource coordinator 712 may include a software design infrastructure ("SDI") management entity for data center 700. Resource coordinator 712 may include hardware, software, or some combination thereof.

[0272] In at least one embodiment, Figure 7As shown, the framework layer 720 may include a job scheduler 732, a configuration manager 734, a resource manager 736, and a distributed file system 738. The framework layer 720 may include a framework that supports the software 732 of the software layer 730 and / or one or more applications 742 of the application layer 740. The software 732 or the application 742 may include web-based service software or applications, such as those provided by Amazon Web Services, Google Cloud, and Microsoft Azure. The framework layer 720 may be, but is not limited to, a free and open source software network application framework, such as Apache Spark, which may utilize the distributed file system 738 for large-scale data processing (e.g., "big data"). TM (hereinafter referred to as "Spark"). In at least one embodiment, the job scheduler 732 may include a Spark driver to facilitate scheduling of workloads supported by various layers of the data center 700. In at least one embodiment, the configuration manager 734 may be able to configure different layers, such as the software layer 730 and the framework layer 720 including Spark and a distributed file system 738 for supporting large-scale data processing. The resource manager 736 can manage cluster or group computing resources mapped to or allocated to support the distributed file system 738 and the job scheduler 732. In at least one embodiment, the cluster or group computing resources may include grouped computing resources 714 at the data center infrastructure layer 710. The resource manager 736 can coordinate with the resource coordinator 712 to manage these mapped or allocated computing resources.

[0273] In at least one embodiment, software 732 included in software layer 730 may include software used by at least a portion of node CRs 716(1)-716(N), grouped computing resources 714, and / or distributed file system 738 of framework layer 720. One or more types of software may include, but are not limited to, Internet web page search software, email virus scanning software, database software, and streaming video content software.

[0274] In at least one embodiment, one or more applications 742 included in the application layer 740 may include one or more types of applications used by at least a portion of the node CRs 716(1)-716(N), the grouped computing resources 714, and / or the distributed file system 738 of the framework layer 720. The one or more types of applications may include, but are not limited to, any number of genomics applications, cognitive computing, and machine learning applications, including training or inference software, machine learning framework software (e.g., PyTorch, TensorFlow, Caffe, etc.), and / or other machine learning applications used in conjunction with one or more embodiments.

[0275] In at least one embodiment, any of the configuration manager 734, resource manager 736, and resource coordinator 712 can implement any number and type of self-modification actions based on any number and type of data acquired in any technically feasible manner. The self-modification actions can relieve the data center operator of the data center 700 from making potentially bad configuration decisions and can avoid underutilized and / or underperforming portions of the data center.

[0276] The data center 700 may include tools, services, software, or other resources for training one or more machine learning models or using one or more machine learning models to predict or infer information according to one or more embodiments described herein. For example, a machine learning model may be trained by calculating weight parameters according to a neural network architecture using the software and computing resources described in this disclosure with respect to the data center 700. In at least one embodiment, by using weight parameters calculated by one or more training techniques, the resources described in this disclosure with respect to the data center 700 may be used to infer or predict information using a trained machine learning model corresponding to one or more neural networks, such as but not limited to those described herein.

[0277] In at least one embodiment, the data center 700 can use a CPU, an application-specific integrated circuit (ASIC), a GPU, an FPGA, and / or other hardware (or virtual computing resources corresponding thereto) to use the above resources to perform training and / or reasoning. In addition, one or more software and / or hardware resources of the present disclosure can be configured as a service to allow users to train or perform information reasoning, such as image recognition, speech recognition, or other artificial intelligence services.

[0278] Example network environment

[0279] A network environment suitable for implementing embodiments of the present disclosure may include one or more client devices, servers, network attached storage (NAS), other backend devices, and / or other device types. The client devices, servers, and / or other device types (e.g., each device) may be configured to communicate with the client device, server, or other device types. Figure 6 The backend device 700 may be implemented on one or more instances of the computing device 600 of the embodiment of the present invention—for example, each device may include similar components, features and / or functions of the computing device 600. In addition, in the case of implementing a backend device (e.g., a server, NAS, etc.), the backend device may be included as part of the data center 700, an example of which is described herein with respect to Figure 7 Describe in more detail.

[0280] The components of the network environment can communicate with each other through the network, which can be wired, wireless, or both. The network can include multiple networks, or a network of multiple networks. For example, the network can include one or more wide area networks (WANs), one or more local area networks (LANs), one or more public networks (such as the Internet and / or the Public Switched Telephone Network (PSTN)), and / or one or more private networks. In the case where the network includes a wireless telecommunications network, components such as base stations, communication towers, or even access points (and other components) can provide wireless connections.

[0281] Compatible network environments may include one or more peer-to-peer network environments (in which case the server may not be included in the network environment), and one or more client-server network environments (in which case one or more servers may be included in the network environment). In a peer-to-peer network environment, the functionality described herein with respect to the server may be implemented on any number of client devices.

[0282] In at least one embodiment, the network environment may include one or more cloud-based network environments, distributed computing environments, combinations thereof, and the like. A cloud-based network environment may include a framework layer, a job scheduler, a resource manager, and a distributed file system implemented on one or more servers, which may include one or more core network servers and / or edge servers. The framework layer may include a framework for supporting software at the software layer and / or one or more applications at the application layer. The software or application may include network-based service software or application programs, respectively. In an embodiment, one or more client devices may use network-based service software or application programs (e.g., by accessing the service software and / or application programs via one or more application programming interfaces (APIs)). The framework layer may be, but is not limited to, a type of free and open source software network application framework, such as a distributed file system that may be used for large-scale data processing (e.g., "big data").

[0283] A cloud-based network environment can provide cloud computing and / or cloud storage that performs any combination of the computing and / or data storage functions (or one or more portions thereof) described herein. Any of these various functions can be distributed across multiple locations from a central or core server (e.g., one or more data centers that can be distributed across states, regions, countries, the world, etc.). If the connection to a user (e.g., a client device) is relatively close to an edge server, the core server can assign at least a portion of the functionality to the edge server. A cloud-based network environment can be private (e.g., limited to a single organization), public (e.g., available to many organizations), and / or a combination thereof (e.g., a hybrid cloud environment).

[0284] The client device may include Figure 6 At least some of the components, features, and functionality of the described example computing device 600. By way of example and not limitation, the client device may be embodied as a personal computer (PC), a laptop computer, a mobile device, a smartphone, a tablet computer, a smart watch, a wearable computer, a personal digital assistant (PDA), an MP3 player, a virtual reality head mounted display, a global positioning system (GPS) or device, a video player, a camera, a surveillance device or system, a vehicle, a vessel, an aircraft, a virtual machine, a drone, a robot, a handheld communication device, a hospital device, a gaming device or system, an entertainment system, an in-vehicle computer system, an embedded system controller, a remote control, an appliance, a consumer electronic device, a workstation, an edge device, any combination of these described devices, or any other suitable device.

[0285] The present disclosure may be described in the general context of machine-usable instructions or computer codes executed by a computer or other machine such as a personal digital assistant or other handheld device, including computer-executable instructions such as program modules. Typically, program modules including routines, programs, objects, components, data structures, etc. refer to codes that perform specific tasks or implement specific abstract data types. The present disclosure may be practiced in a variety of system configurations, including handheld devices, consumer electronics, general-purpose computers, more specialized computing devices, etc. The present disclosure may also be practiced in a distributed computing environment in which tasks are performed by remote processing devices linked through a communication network.

[0286] As used herein, the description of "and / or" about two or more elements should be interpreted as referring to only one element or a combination of elements. For example, "element A, element B and / or element C" may include only element A, only element B, only element C, element A and element B, element A and element C, element B and element C, or elements A, B and C. In addition, "at least one of element A or element B" may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B. Further, "at least one of element A and element B" may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B. In addition, the use of the term "based on" should not be interpreted as "based only on" or "based solely on". On the contrary, a first element "based on" a second element includes a situation where the first element is based on the second element but may also be based on one or more additional elements.

[0287] The subject matter of the present disclosure is described in detail herein to meet statutory requirements. However, the description itself is not intended to limit the scope of the present disclosure. On the contrary, the inventors have contemplated that the claimed subject matter may also be embodied in other ways to include steps different from the steps described herein in conjunction with other current or future technologies or combinations of similar steps. Moreover, although the terms "step" and / or "block" may be used herein to imply different elements of the method employed, these terms should not be interpreted as implying any particular order among or between the various steps disclosed herein, unless the order of the steps is explicitly described.

Claims

1. A method comprising: determining an image position of a representation of a portion of an object in an image corresponding to an environment, the image position being defined in a two-dimensional (2D) image space based on at least image data corresponding to the image; determining one or more predicted image locations corresponding to the image locations of the representation of the portion of the object, the one or more predicted image locations being determined based on at least: Representation of the ground surface in a real-world environment; a vanishing point projected into the image space; as well as said image location of said representation of said portion of said object; comparing one or more ground plane locations of the portion of the object to the one or more predicted image locations, the one or more ground plane locations determined using 2D map data corresponding to a 2D map of the environment projected into the 2D image space; determining a cost based at least on the comparison of the one or more ground plane locations to the one or more predicted image locations; as well as A system is positioned to the 2D map data based at least on the cost.

2. The method according to claim 1, further comprising: The one or more ground plane positions of the portion of the object corresponding to the projected 2D map of the environment are iteratively updated until the determined cost satisfies a cost threshold.

3. The method according to claim 1, wherein: The vanishing point is determined based on at least an orientation of a camera used to generate the image data relative to a gravity vector, the gravity vector indicating a direction of gravity projected into the 2D image space.

4. The method according to claim 2, wherein: The vanishing point is further determined using one or more characteristics of a camera used to generate the image data.

5. The method according to claim 1, wherein: The representation of the one or more predicted image positions comprises a line segment comprising at least the image position of the representation of the portion of the object in the image corresponding to the environment and the vanishing point projected into the image space.

6. The method according to claim 1, wherein: The image data is generated using a single monocular camera.

7. The method according to claim 1, wherein: The one or more ground plane locations determined using the 2D map of the environment projected into the 2D image space are projected using a pinhole projection model.

8. A method comprising: determining an image position of a portion of an object in an image in a two-dimensional 2D image space; determining one or more predicted image locations corresponding to the image location of the portion of the object; comparing one or more ground plane positions of the portion of the object to the one or more predicted image positions, the one or more ground plane positions determined using a 2D map of the environment projected into the 2D image space; as well as The system is positioned to the 2D map based at least on the comparing.

9. The method according to claim 8, wherein: The one or more predicted image positions are determined based on at least: Representation of the ground surface in a real-world environment; a vanishing point projected into the 2D image space; and The image location of the representation of the portion of the object.

10. The method according to claim 8, further comprising: The one or more ground plane positions of the portion of the object corresponding to the projected 2D map of the environment are iteratively updated until a difference between the one or more ground plane positions and the one or more predicted image positions satisfies an error threshold.

11. The method according to claim 9, wherein: The vanishing point is determined based at least on an orientation of a camera used to capture the image relative to a gravity vector indicating a direction of gravity projected into the 2D image space.

12. The method according to claim 11, wherein: The vanishing point is further determined using one or more characteristics of the camera.

13. The method according to claim 9, wherein: The representation of the one or more predicted image positions comprises a line segment comprising at least the image position of the representation of the portion of the object in the image corresponding to the environment and the vanishing point projected into the image space.

14. The method according to claim 8, wherein: The images are captured using a single monocular camera.

15. The method according to claim 8, wherein: The one or more ground plane locations determined using the 2D map of the environment projected into the 2D image space are projected using a pinhole projection model.

16. A processor comprising a processing circuit configured to perform operations comprising: determining an image position of a portion of an object in an image in a two-dimensional 2D image space; determining one or more predicted image locations corresponding to the image location of the portion of the object; comparing one or more ground plane positions of the portion of the object to the one or more predicted image positions, the one or more ground plane positions determined using a 2D map of the environment projected into the 2D image space; as well as The system is positioned to the 2D map based at least on the comparing.

17. The processor of claim 16, wherein: The one or more predicted image positions are determined based on at least: Representation of the ground surface in a real-world environment; a vanishing point projected into the 2D image space; and The image location of the representation of the portion of the object.

18. The processor of claim 16, further comprising: The one or more ground plane positions of the portion of the object corresponding to the projected 2D map of the environment are iteratively updated until a difference between the one or more ground plane positions and the one or more predicted image positions satisfies an error threshold.

19. The processor of claim 17, wherein: The vanishing point is determined based at least on an orientation of a camera used to capture the image relative to a gravity vector indicating a direction of gravity projected into the 2D image space.

20. The processor of claim 16, wherein: The processor is included in a system, the system comprising at least one of the following: control systems for autonomous or semi-autonomous machines; Perception systems for autonomous or semi-autonomous machines; A system for performing simulation operations; Systems for performing digital twin operations; A system for performing light transport simulations; A system for performing collaborative content creation of 3D assets; Systems for performing deep learning operations; a system for presenting at least one of augmented reality content, virtual reality content, or mixed reality content; A system for hosting one or more live streaming applications; Systems implemented using edge devices; Systems implemented using robots; Systems for performing conversational AI operations; A system implementing one or more Large Language Models (LLMs); Systems for performing generative AI operations; Systems for generating synthetic data; A system comprising one or more virtual machines VM; A system implemented at least in part in a data center; or A system implemented at least in part using cloud computing resources.

Images