Feature Position Identification for Autonomous Systems and Applications

Abstract

This paper describes, through various examples, feature position identification for autonomous and semi-autonomous systems and applications. Systems and methods are disclosed that utilize LiDAR data to determine the positions of road markings within an environment. For example, the LiDAR data can be used to generate one or more images, such as a top-down image, a bird's-eye view (BEV) image, and / or an intensity image representing the environment. The image(s) can then be processed using one or more machine learning models configured to determine information associated with the road markings, such as boundary shapes (e.g., boundary boxes) indicating the positions of the road markings within the image(s), classifications associated with the road markings (e.g., the types of road markings), and / or any other information.The information can then be used to perform one or more processes, such as updating a map of the surroundings and / or navigating a vehicle within the surroundings.

Description

BACKGROUND

[0001] For vehicles (e.g., autonomous vehicles, semi-autonomous vehicles, robots, etc.) to operate safely in environments, the vehicles must be able to effectively perform vehicle maneuvers—such as lane keeping, lane changing, lane splitting, turning, stopping and starting at intersections, crosswalks, and the like, and / or other vehicle or machine maneuvers. For example, for a vehicle to navigate through rural roads (e.g., city streets, secondary roads, neighborhood streets, etc.) and on highways (e.g., multi-lane roads), the vehicle must navigate between one or more divisions or boundaries (e.g., lanes, intersections, crosswalks, boundaries, etc.) of a road, which are often marked using road markings—such as arrows, text, graphics, and / or other types of content or features.Therefore, it is important that vehicles are able to detect the road markings within the environments so that vehicles are able to determine how to navigate according to the rules associated with the road markings.

[0002] To detect road markings, vehicles may at least partially use maps that correspond to the environments the vehicles are navigating. For example, the maps may indicate the locations of important features that the vehicles need to identify while navigating, such as road surfaces and road markings. Conventional approaches to determining road marking locations for these maps involve processing image data generated using image sensors from vehicles that have navigated within the environments. For example, the image data may represent images depicting the road markings within the environments. Therefore, the systems are capable of processing the image data, such as using one or more image processing techniques (e.g., object detection, object recognition, etc.), to detect the locations of the road markings within the images.The systems can then use the road marker positions from the images to determine the corresponding road marker positions within the maps.

[0003] Although these systems are capable of determining the positions of road markings within the environments using image processing, there may be room for improving the accuracy of these systems. For example, it may be difficult to obtain very accurate results of road marking positions, such as centimeter-level results, when determining road marking positions using image processing—e.g., because images provide a two-dimensional (2D) representation of the environment, which must then be converted into a three-dimensional (3D) space for localization and navigation. Additional factors can also affect the accuracy of these systems, such as distances to road markings, occlusions within the images, perspective distortions, and scene variations. SUMMARY

[0004] Embodiments of the present disclosure relate to feature location identification for autonomous and semi-autonomous systems and applications. Systems and methods are disclosed that utilize LiDAR data and / or other data types, such as RADAR data, generated using one or more sensors to determine the positions of road markings within an environment. For example, the LiDAR data may be used to generate one or more images, such as one or more top-down images, one or more bird's eye view (BEV) images, and / or one or more intensity images, representing the environment. The image(s) may then be processed using one or more machine learning models configured to determine information associated with the road markings, such as boundary shapes (e.g.,Bounding boxes) indicating the positions of the road markings within the image(s), classifications associated with the road markings (e.g., the types of road markings), and / or any other information. The information can then be used to perform one or more processes, such as updating a map of the environment and / or navigating a vehicle within the environment.

[0005] In contrast to conventional systems, such as those described above, current systems, in some embodiments, are capable of more accurately determining the positions of road markings within environments, such as at the centimeter level. This is because current systems can use LiDAR data processing—and in particular, the image(s) generated using attributes (e.g., intensities, etc.) associated with points represented by the LiDAR data—to determine the positions of the road markings, which can be more accurate than using image processing.Additionally, by generating the image(s) used to determine the positions of the road markings using the LiDAR data, the accuracy of current systems in determining the road markings may, in some embodiments, not be affected by factors such as distances to the road markings, sensor occlusion, perspective distortions, and / or scene variations. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The present systems and methods for feature position identification for autonomous and semi-autonomous systems and applications are described in detail below with reference to the attached drawing figures, where: Fig. 1 illustrates an exemplary data flow diagram for a process for determining information associated with road markings, according to some embodiments of the present disclosure; Fig. 2A-2B illustrate an example of generating an image using LiDAR data according to some embodiments of the present disclosure; Fig. 3 illustrates an example of using one or more machine learning models to process image data to determine information associated with road markings, according to some embodiments of the present disclosure; Fig. 4 illustrates an example output of one or more machine learning models trained to determine information associated with road markings, according to some embodiments of the present disclosure; Fig. 5 illustrates an example of updating a map to include information associated with road markers, according to some embodiments of the present disclosure; Fig. 6 is a data flow diagram illustrating a process for training one or more machine learning models to determine information associated with road markings, according to some embodiments of the present disclosure; Fig. 7-8 illustrate flowcharts showing methods for using LiDAR data to update a map to indicate a position of a road marker within an environment, according to some embodiments of the present disclosure; Fig. 9A is an illustration of an exemplary autonomous vehicle according to some embodiments of the present disclosure; Fig. 9B shows an example of camera positions and fields of view for the exemplary autonomous vehicle of Fig. 9A is in accordance with some embodiments of the present disclosure; Fig. 9C is a block diagram of an example system architecture for the example autonomous vehicle of Fig. 9A is in accordance with some embodiments of the present disclosure; Fig. 9D is a system diagram for communication between one or more cloud-based servers and the example autonomous vehicle of Fig. 9A is in accordance with some embodiments of the present disclosure; Fig. 10 is a block diagram of an exemplary computing device suitable for use in implementing some embodiments of the present disclosure; and Fig. 11 is a block diagram of an exemplary data center suitable for use in implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0007] Systems and methods are disclosed relating to feature location identification for autonomous and semi-autonomous systems and applications. Although the present disclosure is described with respect to an exemplary autonomous vehicle 900 (alternatively referred to herein as "vehicle 900" or "ego machine 900," an example of which is described with respect to the Fig. 9A- Fig. 9D), this is not intended to be limiting. For example, the systems and methods described herein may be used, without limitation, by 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, controlled and uncontrolled robots or robotic platforms, warehouse vehicles, off-highway vehicles, vehicles coupled to one or more trailers, flying vessels, boats, shuttles, emergency response vehicles, motorcycles, electric or motorized bicycles, aircraft, construction vehicles, underwater vehicles, drones, and / or other types of vehicles.Furthermore, although the present disclosure may be described with respect to object or feature detection and / or mapping, this is not intended to be limiting, and the systems and methods described herein may be used in augmented reality, virtual reality, mixed reality, robotics, security and surveillance, autonomous or semi-autonomous machine applications, and / or any other technology spaces where object or feature detection and / or mapping may be used.

[0008] For example, one or more systems may receive sensor data generated using one or more sensors of one or more vehicles, where the sensor data represents at least a portion of an environment. As described herein, the sensor data may include, but is not limited to, image data generated using one or more image sensors (e.g., one or more cameras), LiDAR data generated using one or more LiDAR sensors, RADAR data generated using one or more RADAR sensors, motion data generated using one or more motion sensors (e.g., one or more inertial measurement unit (IMU) sensors), and / or any other types of sensor data generated using any other type of sensor.In some examples, the system(s) may then preprocess at least a portion of the sensor data. For example, the system(s) may use the motion data to align points associated with multiple frames (e.g., multiple rotations of the LiDAR sensor(s)) of the LiDAR data. The system(s) may then use this alignment to generate a point cloud including the points, wherein a density of the points included in the point cloud increases based on the alignment of the multiple frames. As described in more detail herein, by performing such processes to generate the point cloud, the accuracy associated with determining positions of road markings may increase.

[0009] The system(s) may then use at least a portion of the data, such as the point cloud, to generate one or more images associated with the environment. In some examples, the image(s) may include a top-down image (e.g., BEV image), such as a map, representing the environment. In some examples, such as when the image(s) is / are generated using the point cloud, the image(s) (e.g., the top-down image) may represent the intensities associated with points within the environment. Because, in such examples, the intensities of the points vary based on one or more factors, such as the compositions of the surfaces associated with the points (e.g., the colors of the surfaces for which the light is reflected), the image(s) may indicate the structure of the road marking.For example, if the road marking includes an arrow and / or text painted on the road using a particular color, such as white and / or yellow, the image(s) may depict the points associated with the road marking as a different color than the points associated with other features, such as the road itself (which may be black, gray, or the like).

[0010] The system(s) may then process the image(s) using one or more machine learning models trained to determine information associated with objects, such as road markings, represented by the image(s). For example, for a road marking represented by an image, the machine learning model(s) may determine at least a position of the road marking as represented by the image, a position of a bounding shape (e.g., bounding box) within the image associated with the road marking (e.g., enclosing the road marking), a classification associated with the road marking, and / or any other information. In some examples, the position associated with the road marking may include positions of points (e.g., coordinates), such as pixels, representing the road marking within the image(s).Additionally, the boundary shape may be associated with vertices, such as points (e.g., pixels) of the image(s) that define the structure of the boundary shape. In some examples, and as described in more detail herein, the boundary shape may include an orientation based at least on a direction of travel associated with a road for which the pavement marking is positioned. Further, the classification may include a type of pavement marking, such as, but not limited to, a straight arrow, a right arrow, a left arrow, a stop indicator, a yield indicator, a crosswalk indicator, a school zone indicator, and / or any other type of pavement marking.

[0011] The system(s) may then perform one or more processes using the information associated with the road markings. In some examples, for example, the system(s) may update a map associated with the environment using the information—e.g., by encoding the information into the map data. For a road marking, for example, the system(s) may update the map to show the location of the road marking and / or the classification associated with the road marking. In such an example, the system(s) may use one or more techniques to update the map using the image(s). For example, the system(s) may determine the portion of the image(s) associated with the road marking using the boundary shape, such as by using positions (e.g.,Coordinates) of vertices associated with the boundary shape. The system(s) may then determine a portion of the map corresponding to the portion of the image(s), such as by converting the positions associated with the vertices of the boundary shape into positions on the map. Additionally, the system(s) may update the portion of the map to include similar information as the portion of the image(s). For example, the system(s) may update (e.g., convert) points (e.g., pixels) within the portion of the map to be similar to points (e.g., pixels) within the portion of the image(s). The system(s) may then perform similar processes for one or more other road markers identified within the image(s).

[0012] Additionally or alternatively, in some examples, such as when the system(s) is / are executed on a vehicle navigating the environment, the system(s) may use the information associated with the road markings to determine how to navigate. For example, for a road marking, the system(s) may cause the vehicle to navigate according to one or more rules associated with the road marking. For a first example, if the road marking includes a stop indicator painted on the road surface, then the system(s) may cause the vehicle to stop at a location on the road indicated by the stop indicator.For a second example, if the road marking includes a right arrow, again painted on the road surface, then the system(s) may cause the vehicle to turn right as it navigates along the road.

[0013] As described herein, the system(s) may use the machine learning model(s) to process the image(s) to determine the information associated with the road markings. Thus, in some examples, the system(s) may use one or more processes to train the machine learning model(s) to determine the information. For example, the system(s) may train the machine learning model(s) using training data, such as images depicting road markings located within one or more environments. In some examples, the training data is generated similarly to images that are later processed by the machine learning model(s), such as using sensor data (e.g., LiDAR data, etc.) generated using one or more sensors (e.g., one or more LiDAR sensors, etc.) of one or more vehicles.Additionally, the system(s) may train the machine learning model(s) using appropriate ground truth data, such as ground truth data indicating positions of the road markings within the training data, positions of boundary shapes (e.g., bounding boxes) associated with the road markings, classifications associated with the road markings, and / or any other information associated with the road markings. Training the machine learning model(s) is described in more detail herein.

[0014] The systems and methods described herein may be used, without limitation, by 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, controlled and uncontrolled robots or robotic platforms, warehouse vehicles, off-road vehicles, vehicles coupled to one or more trailers, flying vessels, boats, shuttles, emergency response vehicles, motorcycles, electric or motorized bicycles, aircraft, construction vehicles, underwater vehicles, drones, and / or other types of vehicles.Furthermore, the systems and methods described herein may be used for a variety of purposes, including, but not limited to, machine control, machine locomotion, machine driving, synthetic data generation, model training, perception, augmented reality, virtual reality, mixed reality, robotics, security 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 for 3D assets, cloud computing, and / or any other suitable applications.

[0015] Disclosed embodiments may be included in a variety of different systems, such as automotive systems (e.g., a control system for an autonomous or semi-autonomous machine, a perception system for an autonomous or semi-autonomous machine), systems implemented using a robot, air systems, media systems, boat systems, intelligent area monitoring systems, systems for performing deep learning operations, systems for performing simulation operations, systems for performing digital twin operations, systems implemented using an edge device, systems implementing one or more language models—such as one or more large language models (LLMs), systems including one or more virtual machines (VMs), systems for performing synthetic data generation operations, systems,that are at least partially implemented in a data center, systems for performing conversational AI operations, systems for performing light transport simulation, systems for performing collaborative content creation for 3D assets, systems that are at least partially implemented using cloud computing resources, and / or other types of systems.

[0016] With reference to Fig. 1 illustrates Fig. 1 illustrates an exemplary dataflow diagram for a process 100 for determining information associated with road markings, according to some embodiments of the present disclosure. It should be understood that these and other arrangements described herein are set forth as examples only. Other arrangements and elements (e.g., engines, interfaces, functions, sequences, groupings of functions, etc.) may be used in addition to or in place of those shown, and some elements may be omitted entirely. Furthermore, many of the elements described herein are functional entities that may be implemented as discrete or distributed components or in conjunction with other components, and in any suitable combination and location. Various functions described herein as being performed by entities may be performed by hardware, firmware, and / or software.For example, various functions may be performed by a processor executing instructions stored in memory. In some embodiments, the systems, methods, and processes described herein may be implemented using similar components, features, and / or functionality as those of the exemplary autonomous vehicle 900 of FIG. Fig. 9A- Fig. 9D, the exemplary computing device 1000 of Fig. 10 and / or the exemplary data center 1100 of Fig. 11 are executed.

[0017] The process 100 may include an aggregation component 102 that processes LiDAR data 104 to generate point cloud data 106 representing a 3D point cloud. For example, the LiDAR sensor(s) that generated the LiDAR data 104 may include a given frame rate, such as, but not limited to, 10 frames per second (FPS), 15 FPS, 30 FPS, and / or any other frame rate. Thus, to generate the point cloud data 106, the aggregation component 102 may initially utilize motion data 108 generated using one or more motion sensors, such as one or more IMU sensors, representing the motion of the vehicle(s), to align the frame images represented by the LiDAR data 104. The aggregation component 102 can then use the alignment to generate the point cloud data 106.By initially aligning multiple individual images with each other when generating the point cloud data 106, the 3D point cloud represented by the point cloud data 106 may include a dense distribution of points. This may help increase the accuracy and / or precision of determining information associated with road markings within environments, as described herein.

[0018] In some examples, the LiDAR data 104 and / or the point cloud data 106 may represent intensities associated with at least a portion of the points. For example, and as described herein, the LiDAR sensor(s) used to generate the LiDAR data 104 may measure the intensities of the points at the time the light returns to the LiDAR sensor(s). In some examples, the intensities may be represented using a number, such as a number between 0 and 256 (although other ranges may be used in other examples), where the number is based on the composition (e.g., color, texture, material, etc.) of the surface for which the light is being reflected. For example, a low number may indicate low reflectivity, while a high number may indicate high reflectivity.In some examples, the intensity may depend on other factors such as angles of incidence, areas of the spots, moisture content, and / or the like.

[0019] The process 100 may include an image component 110 that processes the point cloud data 106 to generate image data 112. In some examples, the image data 112 may represent one or more images (e.g., one or more maps) that represent the environment associated with the LiDAR data 104 (e.g., the surfaces for which light is reflected within the environment). For example, the image data 112 may represent a top-down image (BEV image) that represents at least one or more surfaces within the environment. In some examples, the image component 110 may generate the image data 112 using the intensities associated with the points, such that the image data 112 represents one or more intensity images that represent the environment.Because in such examples and as described herein, the intensities of the dots may vary based on one or more factors, such as the colors of the surfaces associated with the dots (e.g., the colors of the surfaces for which the light is reflected), the image(s) may indicate the structure of the pavement marking. For example, if the pavement marking includes an arrow painted on the road surface using a particular color, such as white and / or yellow, the image(s) may depict the dots associated with the arrow as a different color than the dots associated with other features, such as the road surface itself.

[0020] For example, illustrate Fig. 2A-2B illustrate an example of generating an image using LiDAR data (e.g., LiDAR data 104) according to some embodiments of the present disclosure. In the example of Fig. 2A, one or more vehicles may navigate around an environment 202 and generate LiDAR data representing the environment 202. As shown, the environment 202 may include at least four different road markings 204(1)-(4) (also referred to individually as a "road marking 204" or in the plural as "road markings 204"), wherein the first road marking 204(1) includes a right arrow located on a first road 206(1) in the environment 202, the second road marking 204(2) includes a straight arrow located on a second road 206(2) within the environment 202, the third road marking 204(3) includes a crosswalk marking located on a third road 206(3) in the environment 202, and the fourth road marking 204(4) includes a stop indicator located on a fourth road 206(4) in the environment 202. Although in the example of Fig. 2A for clarity, the color of the road markings 204 may differ from the color of the roads 206(1)-(4) (also referred to individually as "road 206" or plurally as "roads 206"). For example, the road markings 204 may include a lighter color, such as white, while the roads 206 may include a darker color, such as black.

[0021] Fig. 2B illustrates an example of an image 208, such as an intensity image, that may be generated using the LiDAR data. As described above, the road markings 204 may be painted on the roads 206(1)-(4), for example, using a particular color, such as white, so that the road markings 204 are more reflective and / or visible to drivers. Thus, first points associated with the first road marking 204(1) may be associated with first intensities 210(1), second points associated with the second road marking 204(2) may be associated with second intensities 210(2), third points associated with the third road marking 204(3) may be associated with third intensities 210(3), and fourth points associated with the fourth road marking 204(4) may be associated with fourth intensities 210(4).Additionally, points associated with other surfaces within the environment 202, such as the road surfaces 206, may be associated with different intensities 212. For this reason, the image 208 may represent the points associated with the road markings 204 as including a color that is different from the points associated with the other surfaces.

[0022] Referring again to the example from Fig. 1, the process 100 may include one or more machine learning models 114 that process at least a portion of the image data 112 and, based at least on the processing, generate and / or output object data 116. As shown, the object data 116 may include, at least, position data 118 representing one or more positions of one or more road markers represented by the image(s), boundary data 120 representing one or more boundary shapes (e.g., one or more bounding boxes) indicating one or more portions of the image(s) associated with (e.g., representing) the road marker(s), and classification data 122 representing one or more classifications associated with the road marker(s). In some examples, a position associated with a road marker may include positions (e.g.,coordinates) of points, such as pixels, that represent the road marking within an image. Additionally, a boundary shape associated with a road marking may include vertices, such as points (e.g., pixels) of the image, that define the structure of the boundary shape. In some examples, and as described in more detail herein, the boundary shape may include an orientation based at least on a direction of travel associated with a road for which the road marking is positioned. Further, a classification associated with a road marking may include a type of road marking, such as, but not limited to, a straight arrow, a right arrow, a left arrow, a stop indicator, a yield indicator, a crosswalk indicator, a school zone indicator, and / or any other type of road marking.

[0023] For example, Fig. 3 illustrates an example of one or more machine learning models 302 (which may represent and / or include machine learning model(s) 114) processing image data 304 (which may represent and / or include image data 112) to determine information associated with road markings, according to some embodiments of the present disclosure. The machine learning model(s) 302 may include one or more neural networks that perform one or more of the processes described herein. In some examples, the neural network(s) may include or be referred to as a convolutional neural network and thus may alternatively be referred to herein as a convolutional neural network, convolutional network, or CNN. In some examples, the neural network(s) may include any type of neural network capable of performing the processes described herein.

[0024] The machine learning model(s) 302 may use the image data 304 (with or without preprocessing) as input. The image data 304 may represent one or more images (e.g., one or more maps) representing an environment associated with LiDAR data (e.g., the surfaces for which light is reflected within the environment). For example, the image data 304 may represent a top-down image (BEV image) representing at least one or more surfaces within the environment. In some examples, the image data 304 may be generated using the intensities associated with the points, such that the image data 304 represents one or more intensity images representing the environment. In some examples, the image data 304 may be input as a single image or may be input using batching, such as mini-batching.For example, two or more images can be used as inputs together (e.g., simultaneously).

[0025] The image data 304 may be input to one or more feature extractor layers 306 of the machine learning model(s) 302. The feature extractor layer(s) 306 may include any number of layers, such as layers 306-306C. One or more of the layers 306 may include an input layer. The input layer may contain values ​​associated with the image data 304. For example, if the image data 304 represents an image(s), the input layer may contain values ​​representative of the raw pixel values ​​of the image(s) as a volume (e.g., a width, W, a height, H, and color channels, C (e.g., RGB), such as 32 by 32 by 3) and / or a batch size, B (e.g., where batching is used).

[0026] One or more layers 306 may include convolutional layers. The convolutional layers may compute the output of neurons connected to local regions in an input layer (e.g., the input layer), where each neuron computes a dot product between its weights and a small region to which it is connected in the input volume. An output from a convolutional layer may be a different volume, with one of the dimensions based on the number of filters applied (e.g., the width, height, and number of filters, such as 32 × 32 × 12 if 12 were the number of filters).

[0027] One or more of layers 306 may include a rectified linear unit (ReLU) layer. The ReLU layer(s) may apply an element-wise activation function, such as max(0,x), thresholding at, for example, zero. The resulting volume of a ReLU layer may be the same as the volume of the ReLU layer's input.

[0028] One or more of the layers 306 may include a pooling layer. The pooling layer may perform a downsampling operation along the spatial dimensions (e.g., height and width), which may result in a smaller volume than the input to the pooling layer (e.g., 16 × 16 × 12 of the 32 × 32 × 12 input volume). In some examples, the machine learning model(s) 302 may not include pooling layers. In such examples, other types of convolutional layers may be used instead of pooling layers. In some examples, the feature extractor layer(s) 306 may include alternating convolutional layers and pooling layers.

[0029] One or more of the layers 306 may include a fully connected layer. Each neuron in the fully connected layer(s) may be connected to each of the neurons in the previous volume. The fully connected layer may compute class scores, and the resulting volume may be 1 × 1 × N (where N is a number of classes). In some examples, the feature extractor layer(s) 306 may include a fully connected layer, while in other examples, the fully connected layer of the machine learning model(s) 302 may be the fully connected layer separate from the feature extractor layer(s) 306. In some examples, the feature extractor layer(s) 306 and / or the machine learning model(s) 302 as a whole may not utilize fully connected layers to increase processing times and reduce computational resource requirements.In such examples where fully connected layers are not used, the machine learning model(s) 302 may be referred to as a fully convolutional network.

[0030] One or more of the layers 306 may include deconvolution layer(s) in some examples. However, the use of the term deconvolution may be misleading and is not intended to be limiting. For example, the deconvolution layer(s) may alternatively be referred to as transposed convolutional layers or fractionally shifted convolutional layers. The deconvolution layer(s) may be used to upsample the output of a previous layer. For example, the deconvolution layer(s) may be used to upsample to a spatial resolution equal to the spatial resolution of the input images for the machine learning model(s) 302 or used to upsample to the input spatial resolution of a next layer.

[0031] Although input layers, convolutional layers, pooling layers, ReLU layers, deconvolutional layers, and fully connected layers are discussed herein with respect to the feature extractor layer(s) 306, this is not intended to be limiting. For example, additional or alternative layers 306 may be used in the feature extractor layer(s) 306, such as normalization layers, SoftMax layers, and / or other layer types.

[0032] The output of the feature extractor layer(s) 306 may be an input to one or more information layers 308. The information layer(s) 308-C may use one or more of the layer types described herein with respect to the feature extractor layer(s) 306. As described herein, in some examples, the information layer(s) 308 may not include fully connected layers to reduce processing speeds and reduce computational resource requirements. In such examples, the information layer(s) 308 may be referred to as fully convolutional layers.

[0033] Depending on the embodiment, different orders and numbers of layers 306 and 308 of the machine learning model(s) 302 may be used. For example, if two or more cameras or other sensor types are used to generate inputs, there may be a different order and number of layers 306 and 308. As another example, a different order and numbering of layers may be used depending on the type of sensor used to generate the image data 304 or the type of image data 304 (e.g., RGB, YUV, etc.). Thus, the order and number of layers 306 and 308 of the machine learning model(s) 302 is not limited to any one architecture.

[0034] Additionally, some of the layers 306 and 308 may include parameters (e.g., weights and / or biases)—such as the feature extractor layer(s) 306 and / or the information layer(s) 308—while others may not, such as the ReLU layers and pooling layers. In some examples, the parameters may be learned by the machine learning model(s) 302 during training. Furthermore, some of the layers 306 and 308 may include additional hyperparameters (e.g., learning rate, step size, epochs, kernel size, number of filters, type of pooling for pooling layers, etc.)—such as the convolutional layer(s), the deconvolutional layer(s), and the pooling layer(s)—while other layers may not, such as the ReLU layer(s). Various activation functions can be used, including ReLU, leaky ReLU, sigmoid, hyperbolic tangent (tan h), exponential linear unit (ELU), etc.The parameters, hyperparameters and / or activation functions are not intended to be limited and may vary depending on the implementation.

[0035] In any example, the output of the machine learning model(s) 302 may include object data 310 (which may represent and / or include the object data 116).

[0036] Fig. 4 illustrates an example output 402 of the machine learning model(s) 114 (and / or the machine learning model(s) 302) trained to determine information associated with road markings, according to some embodiments of the present disclosure. As shown, the output 402 may include at least boundary shapes 404(1)-(4) (also referred to individually as a "boundary shape 404" or plurally as "boundary shapes 404") associated with the road markings 204.For example, the output 402 may include the first bounding shape 404(1) indicating a first portion of the image 208 associated with the first road marking 204(1), the second bounding shape 404(2) indicating a second portion of the image 208 associated with the second road marking 204(2), the third bounding shape 404(3) indicating a third portion of the image 208 associated with the third road marking 204(3), and the fourth bounding shape 404(4) indicating a fourth portion of the image 208 associated with the fourth road marking 204(4). Although the example of . Fig. While FIG. 4 illustrates the boundary shapes 404 as including boxes (e.g., rectangles), in other examples, boundary shapes may include any other type of shape. For example, boundary shapes may include circles, triangles, pentagons, hexagons, octagons, and / or any other shape.

[0037] In some examples, the bounding shapes 404 may include vertices and / or be defined using vertices. For example, and for the first bounding shape 404(1) (and / or similarly for the other bounding shapes 404(2)-(4), which are not illustrated for clarity), the first bounding shape 404(1) may include vertices 406(1)-(4) (also referred to individually as a "vertex 406" or plurally as "vertices 406"). In some examples, a vertex 406 may be associated with a position within the image 208, such as coordinates (e.g., an x-coordinate, a y-coordinate, etc.) indicating the position of the vertex within the image 208.Accordingly, and as described herein, the actual output from the machine learning model(s) 114 may include the positions of the vertices 406 and / or the positions of the vertices 406 may be used when updating a map associated with the environment 202 to include information associated with the first road marker 204(1).

[0038] In some examples and as shown by the example of Fig. 4, the boundary shapes 404 may include orientations based at least on directions of travel associated with the roads 206 for which the road markings 204 are positioned. For a first example, the third boundary shape 404(3) may be oriented based at least on a direction of travel 408(1) associated with the third road 206(3) for which the third road marking 204(3) is positioned. For example, and as shown, the third boundary shape 404(3) is oriented orthogonal to the direction of travel 408(1) such that one side of the third boundary shape 404(3) is perpendicular to the direction of travel 408(1). For a second example, the fourth boundary shape 404(4) may be oriented based at least on a direction of travel 408(2) associated with the fourth road 206(4) for which the fourth road marking 204(4) is positioned.For example, and as shown, the fourth boundary shape 404(4) is oriented orthogonally to the direction of travel 408(2), such that one side of the fourth boundary shape 404(4) is perpendicular to the direction of travel 408(2). In some examples, the boundary shapes 404 may include such orientations to better define the portions of the image 208 associated with the road markings 204, since the road markings 204 also tend to be oriented based on the directions of travel of the roads 206.

[0039] Referring again to the example from Fig. 1, the process 100 may include a map component 124 that uses the object data 116 to generate and / or update a map associated with the environment, where the map is represented by map data 126. For example, the map component 124 may update the map to include information associated with the road markers, such as, but not limited to, the positions of the road markers within the environment and / or the classifications associated with the road markers. In some examples, the map component 124 may use the boundary shapes associated with the road markers to update the positions of the road markers within the map. For example, and for a road marker, the map component 124 may update the position of the boundary shape (e.g., the positions of the vertices, the positions of pixels within the boundary shape, etc.).) within an image to determine a corresponding position (e.g., corresponding pixels) of the map. The map component 124 can then convert the corresponding position of the map to be similar to the position of the image. In other words, the map component 124 can convert a portion of the map representing a similar part of the environment as the boundary shape from the image to be similar to the portion of the image.

[0040] For example, Fig. 5 illustrates an example visualization for updating a map 502 to include information associated with road markers 204, according to some embodiments of the present disclosure. As shown, map 502 may represent at least the environment 202 that includes road marker 204 associated with roads 206. Thus, map component 124 may perform one or more processes to determine that a first portion 504(1) of map 502 corresponds to first boundary shape 404(1), a second portion 504(2) of map 502 corresponds to second boundary shape 404(2), a third portion 504(3) of map 502 corresponds to third boundary shape 404(3), and a fourth portion 504(4) of map 502 corresponds to fourth boundary shape 404(4). In some examples, the map component 124 makes the determinations using the vertices associated with the boundary shapes 404.For example, and with respect to the first boundary shape 404(1), the map component 124 may determine that the vertices 406 associated with the boundary shape 404 correspond to points 506(1)-(4) (e.g., pixels) of the map 502. The map component 124 may then determine the first portion 504(1) using the points 506(1)-(4).

[0041] The map component 124 may then use the correspondences to update the map 502. For example, and as shown, the map component 124 may update (e.g., transform) the first portion 504(1) of the map 502 to be similar to the portion of the image 208 associated with the first boundary shape 404(1), such that the first portion 504(1) of the map 502 includes a first representation 508(1) of the first road marker 204(1). The map component 124 may also update (e.g., transform) the second portion 504(2) of the map 502 to be similar to the portion of the image 208 associated with the second boundary shape 404(2), such that the second portion 504(2) of the map 502 includes a second representation 508(2) of the second road marker 204(2). Additionally, the map component 124 may update (e.g.,convert) to be similar to the portion of the image 208 associated with the third boundary shape 404(3), such that the third portion 504(3) of the map 502 includes a third representation 508(3) of the third road marker 204(3). Further, the map component 124 may update (e.g., convert) the fourth portion 504(4) of the map 502 to be similar to the portion of the image 208 associated with the fourth boundary shape 404(4), such that the fourth portion 504(4) of the map 502 includes a fourth representation 508(4) of the fourth road marker 204(4).

[0042] Referring again to the example from Fig. 1, in some examples, the process 100 may include performing one or more additional and / or alternative processes using the object data 116. For example, a vehicle navigating within the environment may use the object data 116 to determine one or more operations for navigating. For example, the vehicle may use the object data 116 to determine at least the position of the road marking within the environment and the classification associated with the road marking. The vehicle may then navigate based on the position and / or the classification, such as by following one or more rules associated with the road marking.

[0043] With reference to Fig. 6 is Fig. 6 is a data flow diagram illustrating a process 600 for training the machine learning model(s) 114 (and / or similarly the machine learning model(s) 302) to determine information associated with road markings, according to some embodiments of the present disclosure. As shown, the machine learning model(s) 114 may be trained using image data 602 (e.g., training image data). The image data 602 may represent one or more images (e.g., one or more maps) representing an environment associated with LiDAR data (e.g., the surfaces for which the light is reflected within the environment), similar to the image data 112. For example, the image data 602 may represent a top-down image (BEV image) representing at least one or more surfaces within the environment.In some examples, the image data 602 may be generated using the intensities associated with the points, such that the image data 602 represents one or more intensity images representing the environment. For example, the image data 602 may represent an image similar to the image 208 of the example of FIG. Fig. 2B is similar.

[0044] The machine learning model(s) 114 may be trained using the training image data 602 and corresponding ground truth data 604. The ground truth data 604 may include annotations, labels, masks, and / or the like. In some embodiments, the ground truth data 604 may include, for example, at least positions 606 of road markers within the image(s), boundary shapes 608 associated with the road markers within the image(s), and / or classifications 610 associated with the road markers within the image(s). The ground truth data 604 may be generated within a drawing program (e.g., an annotation program), a computer-aided design (CAD) program, a labeling program, another type of program suitable for generating the ground truth data 604, and / or may be hand-drawn in some examples.In any example, the ground truth data 604 may be synthetically produced (e.g., generated from computer models or renderings), genuinely produced (e.g., designed and produced from real-world data), machine-automated (e.g., using feature analysis and learning to extract features from data and then generate labels), human-annotated (e.g., a labeler or annotator defines the position of the labels), and / or a combination thereof (e.g., a human identifies vertices of polylines, machine generates polygons using polygon rasters). In some examples, corresponding ground truth data 604 may exist for each input image.

[0045] A training engine 612 may use one or more loss functions that measure a loss (e.g., an error) in outputs 614 (which may include object data similar to object data 116) compared to ground truth data 604. Any type of loss function may be used, such as cross-entropy loss, mean squared error, mean absolute error, mean bias error, and / or other loss function types. In some examples, different outputs 614 may have different loss functions. For example, the position outputs may have a first loss function, the boundary shape outputs may have a second loss function, and / or the classification outputs may have a third loss function.In such examples, the loss functions may be combined to form an overall loss, and the overall loss may be used to train the machine learning model(s) 114 (e.g., update its parameters). In any example, backpass calculations may be performed to recursively calculate gradients of the loss function(s) with respect to training parameters. In some examples, weights and biases of the machine learning model(s) 114 may be used to calculate these gradients.

[0046] Now with reference to the Fig. 7 and Fig. 8, each block of the methods 700 and 800 described herein comprises a computational process that may be performed using any combination of hardware, firmware, and / or software. For example, various functions may be performed by a processor executing instructions stored in a memory. The methods 700 and 800 may also be embodied as computer-usable instructions stored on computer storage media. The methods 700 and 800 may be provided by a standalone application, a service, or a hosted service (standalone or in combination with another hosted service), or a plug-in for another product, to name a few. Additionally, the methods 700 and 800 are described by way of example with respect to Fig. 1. However, these methods 700 and 800 may additionally or alternatively be performed by any system or combination of systems, including, but not limited to, those described herein.

[0047] Fig. 7 illustrates a flowchart showing a first method 700 for using LiDAR data to update a map to indicate a position of a road marker within an environment, according to some embodiments of the present disclosure. The method 700 may include, at block B702, generating, based at least on LiDAR data, an image representative of at least a portion of an environment. For example, the image component 110 may process the point cloud data 106, wherein the point cloud data 106 is generated using the LiDAR data 104. As described herein, in some examples, the LiDAR data 104 may be generated using one or more LiDAR sensors of one or more machines navigating the environment. Based at least on the processing, the image component 110 may generate the image data 112 representing the image.As described herein, in some examples, the image may include a top-down image (BEV image) depicting the at least one portion of the environment. Additionally or alternatively, in some examples, the image may include an intensity image depicting the at least one portion of the environment.

[0048] The method 700 may include, at block B704, determining, using one or more machine learning models and based at least on the image, a position associated with a road marking within the environment. For example, the machine learning model(s) 114 may process the image data 112 representing at least the image. Based at least on the processing, the machine learning model(s) 114 may output the object data 116 associated with the road marking. As described herein, the object data 116 may include at least the position data 118 representing the position of the road marking within the image, the boundary data 120 representing the boundary shape within the image associated with the road marking, and the classification data 122 representing the classification associated with the road marking.

[0049] The method 700 may include, at block B706, causing a map to display the position associated with the road marker within the environment. For example, the map component 124 may use the object data 116 to update the map, where the map is represented by the map data 126. As described herein, the map component 124 may update the map to display at least one of the position of the road marker within the environment and the classification associated with the road marker. For example, the map component 124 may use the boundary shape associated with the road marker from the image to convert a corresponding portion of the map to include a representation of the road marker.

[0050] Fig. 8 illustrates a flowchart showing a second method 800 for using LiDAR data to update a map to indicate a position of a road marker within an environment, according to some embodiments of the present disclosure. The method 800 may include, at block B802, generating, based at least on LiDAR data, an image representative of at least a portion of an environment. For example, the image component 110 may process the point cloud data 106, wherein the point cloud data 106 is generated using the LiDAR data 104. As described herein, in some examples, the LiDAR data 104 may be generated using one or more LiDAR sensors of one or more machines navigating the environment. Based at least on the processing, the image component 110 may generate the image data 112 representing the image.As described herein, in some examples, the image may include a top-down image (BEV image) depicting the at least one portion of the environment. Additionally or alternatively, in some examples, the image may include an intensity image depicting the at least one portion of the environment.

[0051] The method 800 may include, at block B804, determining, using one or more machine learning models and based at least on the image, that a portion of the image represents a road marking. For example, the machine learning model(s) 114 may process the image data 112 representing at least the image. Based at least on the processing, the machine learning model(s) 114 may output the object data 116 associated with the road marking. As described herein, the object data 116 may include at least the position data 118 representing a first portion (e.g., first pixels) of the image representing the road marking and / or the boundary data 120 representing a second portion (e.g., second pixels) of the image associated with the road marking.In some examples, the object data 116 may further include the classification data 122 representing the classification associated with the road marking.

[0052] The method 800 may include, at block B806, determining that the portion of the image corresponds to a portion of a map. For example, the map component 124 may use the object data 116 to determine that the portion of the image corresponds to the portion of the map, where the map is represented by the map data 126. In some examples, the map component 124 makes the determination using one or more locations associated with the portion of the image, such as one or more pixel locations and / or one or more coordinates. For example, the map component 124 may determine that the one or more pixel locations from the image correspond to one or more pixel locations from the map.

[0053] The method 800 may include, at block B808, updating the portion of the map to represent the road marker. The map component 124 may then, for example, update the portion of the map to represent the road marker. In some examples, updating the map may include converting one or more pixels associated with the portion of the map to be similar to one or more pixels associated with the image. Example autonomous vehicle

[0054] Fig. 9A is an illustration of an exemplary autonomous vehicle 900 according to some embodiments of the present disclosure. The autonomous vehicle 900 (alternatively referred to herein as "vehicle 900") may include, without limitation, a passenger vehicle, such as a car, a truck, a bus, a first responder vehicle, a shuttle, an electric or motorized bicycle, a motorcycle, a fire engine, a police vehicle, an ambulance, a boat, a construction vehicle, an underwater vehicle, a robotic vehicle, a drone, an aircraft, a vehicle coupled to a trailer (e.g., a semi-trailer truck used to haul cargo), and / or another type of vehicle (e.g., that is unmanned and / or accommodates one or more passengers).Autonomous vehicles are generally described in terms of 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, published June 15, 2018, Standard No. J3016-201609, published September 30, 2016, and previous and future versions of this standard). The vehicle 900 may be capable of operating according to one or more of Level 3 - Level 5 autonomous driving levels. The vehicle 900 may be capable of operating according to one or more of Level 1 - Level 5 autonomous driving levels.For example, the vehicle 900 may be capable of supporting driver assistance (Level 1), partial automation (Level 2), conditional automation (Level 3), high automation (Level 4), and / or full automation (Level 5), depending on the embodiment. The term "autonomous" as used herein may include any and / or all types of autonomy for the vehicle 900 or other machine, such as fully autonomous, highly autonomous, conditionally autonomous, partially autonomous, providing assistive autonomy, semi-autonomous, primarily autonomous, or another designation.

[0055] The vehicle 900 may include components such as a chassis, a vehicle body, wheels (e.g., 2, 4, 6, 8, 18, etc.), tires, axles, and other components of a vehicle. The vehicle 900 may include a propulsion system 950, such as an internal combustion engine, a hybrid electric power plant, a pure electric motor, and / or another type of propulsion system. The propulsion system 950 may be connected to a drivetrain of the vehicle 900, which may include a transmission to enable propulsion of the vehicle 900. The propulsion system 950 may be controlled in response to receiving signals from the throttle / accelerator pedal 952.

[0056] A steering system 954, which may include a steering wheel, may be used to steer the vehicle 900 (e.g., along a desired path or route) when the propulsion system 950 is operating (e.g., when the vehicle is moving). The steering system 954 may receive signals from a steering actuator 956. The steering wheel may be optional for full automation (Level 5) functionality.

[0057] The brake sensor system 946 may be used to apply the vehicle brakes in response to receiving signals from the brake actuators 948 and / or brake sensors.

[0058] The controller(s) 936, which may comprise one or more system-on-chips (SoCs) 904 ( Fig. 9C) and / or GPU(s) may provide signals (e.g., representative of commands) to one or more components and / or systems of the vehicle 900. For example, the controller(s) may send signals to apply the vehicle brakes via one or more brake actuators 948, to apply the steering system 954 via one or more steering actuators 956, to apply the propulsion system 950 via one or more throttles / gas pedals 952. The controller(s) 936 may include one or more on-board (e.g., integrated) computing devices (e.g., supercomputers) that process sensor signals and issue operational commands (e.g., signals representing commands) to enable autonomous driving and / or to assist a human driver in driving the vehicle 900.The controller(s) 936 may include a first controller 936 for autonomous driving functions, a second controller 936 for functional safety functions, a third controller 936 for artificial intelligence functionality (e.g., computer vision), a fourth controller 936 for infotainment functionality, a fifth controller 936 for redundancy in emergency conditions, and / or other controllers. In some examples, a single controller 936 may handle two or more of the foregoing functionalities, two or more controllers 936 may handle a single functionality, and / or any combination thereof.

[0059] The controller(s) 936 may provide the signals to control one or more components and / or systems of the vehicle 900 in response to sensor data received from one or more sensors (e.g., sensor inputs). The sensor data may, for example and without limitation, be provided by the Global Navigation Satellite System (“GNSS”) sensor(s) 958 (e.g., Global Positioning System sensor(s), RADAR sensor(s) 960, ultrasonic sensor(s) 962, LiDAR sensor(s) 964, Inertial Measurement Unit (IMU) sensor(s) 966 (e.g., accelerometer, gyroscope, magnetic compass, magnetometer, etc.), microphone(s) 996, stereo camera(s) 968, wide-view camera(s) 970 (e.g., fisheye cameras), infrared camera(s) 972, surround camera(s) 974 (e.g., 360-degree cameras), long-range and / or medium-range camera(s) 998, speed sensor(s) 944 (e.g., for measuring the speed of the vehicle 900), vibration sensor(s) 942, steering sensor(s) 940, brake sensor(s) (e.g.B. as part of the brake sensor system 946) and / or other sensor types.

[0060] One or more of the controller(s) 936 may receive inputs (e.g., represented by input data) from an instrument cluster 932 of the vehicle 900 and provide outputs (e.g., represented by output data, display data, etc.) via a human-machine interface (HMI) display 934, an audible annunciator, a speaker, and / or via other components of the vehicle 900. The outputs may include information such as vehicle speed, velocity, time, map data (e.g., the high-resolution ("HD") map 922 of Fig. 9C), location data (e.g., the location of the vehicle 900, such as on a map), direction, location of other vehicles (e.g., an occupancy grid), information about objects and status of objects as perceived by the controller(s) 936, etc. For example, the HMI display 934 may display information about the presence of one or more objects (e.g., a road sign, warning sign, traffic light change, etc.) and / or information about maneuvers the vehicle has performed, is performing, or will perform (e.g., changing lanes now, entering exit 34B in two miles, etc.).

[0061] The vehicle 900 further includes a network interface 924 that may utilize one or more wireless antenna(s) 926 and / or modem(s) to communicate over one or more networks. For example, the network interface 924 may be capable of communicating over Long-Term Evolution ("LTE"), Wideband Code Division Multiple Access ("WCDMA"), Universal Mobile Telecommunications System ("UMTS"), Global System for Mobile Communication ("GSM"), IMT-CDMA Multi-Carrier ("CDMA2000"), etc. The wireless antenna(s) 926 may also enable communication between surrounding objects (e.g., vehicles, mobile devices, etc.) using local area network(s), such as Bluetooth, Bluetooth Low Energy ("LE"), Z-Wave, ZigBee, etc., and / or low-power wide area network(s) ("LPWANs"), such as LoRaWAN, SigFox, etc.

[0062] Fig. 9B shows an example of camera positions and fields of view for the example autonomous vehicle 900 of Fig. 9A, according to some embodiments of the present disclosure. The cameras and respective fields of view are an example embodiment and are not intended to be limiting. For example, additional and / or alternative cameras may be included and / or the cameras may be positioned at various locations on the vehicle 900.

[0063] The camera types for the cameras may include, but are not limited to, digital cameras that may be adapted for use with the components and / or systems of the vehicle 900. The camera(s) may operate at Automotive Safety Integrity Level (ASIL) B and / or another ASIL. The camera types may be capable of any image capture rate, such as 60 frames per second (fps), 120 fps, 240 fps, etc., depending on the embodiment. The cameras may be capable of using rolling shutters, global shutters, another type of shutter, or a combination thereof.In some examples, the color filter array may include a red-clear-clear-clear (RCCC) color filter array, a red-clear-clear-blue (RCCB) color filter array, a red-blue-green (RBGC) color filter array, a Foveon X3 color filter array, a Bayer sensor color filter array (RGGB) color filter array, a monochrome sensor color filter array, and / or another type of color filter array. In some embodiments, clear-pixel cameras, such as cameras with an RCCC, an RCCB, and / or an RBGC color filter array, may be used to increase light sensitivity.

[0064] In some examples, one or more of the cameras may be used to perform advanced driver assistance system (ADAS) functions (e.g., as part of a redundant or fail-safe design). For example, a multifunction mono camera may be installed to provide functions including lane departure warning, traffic sign support, and intelligent headlight control. One or more of the cameras (e.g., all cameras) may record and provide image data (e.g., video) simultaneously.

[0065] One or more of the cameras may be mounted in a mounting assembly, such as a custom-designed (three-dimensional ("3D") printed) assembly, to cut out stray light and reflections from inside the car (e.g., reflections from the dashboard reflected in the windshield mirrors) that may interfere with the camera's image data collection capabilities. With reference to wing mirror mounting assemblies, the wing mirror assemblies may be custom 3D printed so that the camera mounting plate conforms to the shape of the wing mirror. In some examples, the camera(s) may be integrated into the wing mirror. For side-view cameras, the camera(s) may also be integrated into the four pillars at each corner of the cabin.

[0066] Cameras with a field of view that includes portions of the environment in front of the vehicle 900 (e.g., forward-facing cameras) can be used for surround visibility to help identify forward paths and obstacles, as well as to help provide, with the aid of one or more controllers 936 and / or control SoCs, information critical for generating an occupancy grid and / or determining preferred vehicle paths. Forward-facing cameras can be used to perform many of the same ADAS functions as LiDAR, including emergency braking, pedestrian detection, and collision avoidance. Forward-facing cameras can also be used for ADAS functions and systems, including lane departure warnings ("LDW"), autonomous cruise control ("ACC"), and / or other functions such as traffic sign recognition.

[0067] A variety of cameras may be used in a forward-facing configuration, including, for example, a monocular camera platform that includes a complementary metal oxide semiconductor ("CMOS") color imager. Another example may be a wide-view camera(s) 970 that may be used to perceive objects coming into view from the periphery (e.g., pedestrians, crossing traffic, or bicycles). Although in Fig. 9B illustrates only one wide-view camera, there may be any number (including zero) of wide-view cameras 970 on the vehicle 900. Additionally, any number of long-range camera(s) 998 (e.g., a long-range stereo camera pair) may be used for depth-based object detection, particularly for objects for which a neural network has not yet been trained. The long-range camera(s) 998 may also be used for object detection and classification, as well as basic object tracking.

[0068] Any number of stereo cameras 968 may also be included in a forward-facing configuration. In at least one embodiment, one or more of the stereo camera(s) 968 may include an integrated control unit comprising a scalable processing unit that may provide programmable logic ("FPGA") and a multi-core microprocessor with an integrated Controller Area Network ("CAN") or an Ethernet interface on a single chip. Such a unit may be used to generate a 3D map of the vehicle's surroundings, including a distance estimate for all points in the image. An alternative stereo camera(s) 968 may include one or more compact stereo vision sensors that may include two camera lenses (one each on the left and right) and an image processing chip that measures the distance from the vehicle to the target object and processes the generated information (e.g.,metadata) to activate the autonomous emergency braking and lane departure warning functions. Other types of stereo camera(s) 968 may be used in addition to or alternatively to those described here.

[0069] Cameras with a field of view that includes portions of the environment to the side of the vehicle 900 (e.g., side view cameras) may be used for surround visibility, providing information used to create and update the occupancy grid, as well as to generate side impact collision warnings. For example, the surround camera(s) 974 (e.g., four surround cameras 974, as shown in Fig. 9B) may be positioned on the vehicle 900. The surround camera(s) 974 may include wide-view camera(s) 970, fisheye camera(s), 360-degree camera(s), and / or the like. For example, four fisheye cameras may be positioned at the front, rear, and sides of the vehicle. In an alternative arrangement, the vehicle may utilize three surround camera(s) 974 (e.g., left, right, and rear) and may utilize one or more other cameras (e.g., a forward-facing camera) as a fourth surround-view camera.

[0070] Cameras with a field of view that includes portions of the environment behind the vehicle 900 (e.g., rearview cameras) may be used for parking assistance, surround view, rear collision warnings, and creating and updating the occupancy grid. A wide variety of cameras may be used, including, but not limited to, cameras that are also suitable as forward-facing cameras (e.g., long-range and / or medium-range camera(s) 998, stereo camera(s) 968), infrared camera(s) 972, etc.), as described herein.

[0071] Fig. 9C is a block diagram of an example system architecture for the example autonomous vehicle 900 of Fig. 9A according to some embodiments of the present disclosure. It should be understood that these and other arrangements described herein are set forth only as examples. Other arrangements and elements (e.g., engines, interfaces, functions, sequences, groupings of functions, etc.) may be used in addition to or in place of those shown, and some elements may be omitted entirely. Furthermore, many of the elements described herein are functional entities that may be implemented as discrete or distributed components or in conjunction with other components, and in any suitable combination and location. Various functions described herein as being performed by entities may be performed by hardware, firmware, and / or software. For example, various functions may be performed by a processor executing instructions stored in memory.

[0072] Each of the vehicle’s components, features and systems 900 in Fig. 9C are illustrated as being connected via bus 902. Bus 902 may include a Controller Area Network (CAN) data interface (alternatively referred to herein as a "CAN bus"). A CAN may be a network within vehicle 900 used to assist in controlling various features and functionality of vehicle 900, such as brake application, acceleration, braking, steering, windshield wipers, etc. A CAN bus may be configured to have dozens 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 revolutions per minute (RPMs), button positions, and / or other vehicle status indicators. The CAN bus may be ASIL B compliant.

[0073] Although bus 902 is described herein as a CAN bus, this is not intended to be limiting. For example, FlexRay and / or Ethernet may be used in addition to or alternatively to the CAN bus. Additionally, although a single wire is used to represent bus 902, this is not intended to be limiting. For example, there may be any number of buses 902, 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 a different protocol. In some examples, two or more buses 902 may be used to perform different functions and / or may be used for redundancy. For example, a first bus 902 may be used for collision avoidance functionality and a second bus 902 may be used for actuation control.In any example, each bus 902 may communicate with any of the components of the vehicle 900, and two or more buses 902 may communicate with the same components. In some examples, each SoC 904, controller 936, and / or computer within the vehicle may have access to the same input data (e.g., inputs from sensors of the vehicle 900) and may be connected to a common bus, such as the CAN bus.

[0074] The vehicle 900 may include one or more controllers 936, such as those described herein with respect to Fig. 9A. The controller(s) 936 may be used for a variety of functions. The controller(s) 936 may be coupled to any of the various other components and systems of the vehicle 900 and may be used to control the vehicle 900, artificial intelligence of the vehicle 900, infotainment for the vehicle 900, and / or the like.

[0075] The vehicle 900 may include one or more systems on a chip (SoC) 904. The SoC 904 may include CPU(s) 906, GPU(s) 908, processor(s) 910, cache(s) 912, accelerators 914, data storage 916, and / or other unillustrated components and features. The SoC(s) 904 may be used to control the vehicle 900 in a variety of platforms and systems. For example, the SoC(s) 904 may be combined in a system (e.g., the system of the vehicle 900) with an HD map 922 that may receive map updates and / or updates via a network interface 924 from one or more servers (e.g., server 978 from Fig. 9D).

[0076] The CPU(s) 906 may include a CPU cluster or CPU complex (alternatively referred to herein as a "CCPLEX"). The CPU(s) 906 may include multiple cores and / or L2 caches. For example, in some embodiments, the CPU(s) 906 may include eight cores in a coherent multiprocessor configuration. In some embodiments, the CPU(s) 906 may include four dual-core clusters, each cluster having a dedicated L2 cache (e.g., a 2 MB L2 cache). The CPU(s) 906 (e.g., the CCPLEX) may be configured to support concurrent cluster operation, allowing any combination of the clusters of the CPU(s) 906 to be active at any one time.

[0077] The CPU(s) 906 may implement power management capabilities that include one or more of the following features: Individual hardware blocks may be automatically clocked when idle to conserve dynamic power; Each core clock may be clocked when the core is not actively executing instructions due to the execution of WFI / WFE instructions; Each core may be independently clocked; Each core cluster may be independently clocked when all cores are clocked or unclocked; and / or Each core cluster may be independently clocked when all cores are clocked. The CPU(s) 906 may further implement an advanced algorithm for managing power states, where allowable power states and expected wake-up times are specified, and the hardware / microcode determines the best power state to enter for the core, cluster, and CCPLEX.The processing cores can support simplified power state input sequences in software, offloading the work to microcode.

[0078] The GPU(s) 908 may include an integrated GPU (alternatively referred to herein as an "iGPU"). The GPU(s) 908 may be programmable and may be efficient for parallel workloads. In some examples, the GPU(s) 908 may use an extended tensor instruction set. The GPU(s) 908 may include one or more streaming microprocessors, where each streaming microprocessor may include an L1 cache (e.g., an L1 cache with a memory capacity of at least 96 KB) and two or more of the streaming microprocessors may share an L2 cache (e.g., an L2 cache with a memory capacity of 512 KB). In some embodiments, the GPU(s) 908 may include at least eight streaming microprocessors. The GPU(s) 908 may use one or more Compute Application Programming Interface(s) (API(s)).Additionally, the GPU(s) 908 may utilize one or more parallel computing platforms and / or programming models (e.g., NVIDIA's CUDA).

[0079] The GPU(s) 908 may be performance-optimized for best performance in automotive and embedded use cases. For example, the GPU(s) 908 may be fabricated on a fin field-effect transistor (FinFET). However, this is not intended to be limiting, and the GPU(s) 908 may be fabricated using other semiconductor manufacturing processes. Each streaming microprocessor may include a number of mixed-precision processing cores partitioned into multiple blocks. For example, and without limitation, 64 PF32 cores and 32 PF64 cores may be partitioned into four processing blocks. In such an example, each processing block may be assigned 16 FP32 cores, 8 FP64 cores, 16 INT32 cores, two NVIDIA mixed-precision TENSOR COREs for deep learning matrix arithmetic, an L0 instruction cache, a warp scheduler, a dispatch unit, and / or a 64 KB register file.Additionally, streaming microprocessors can include independent parallel integer and floating-point data paths to provide efficient execution of workloads with a mix of computation and addressing calculations. Streaming microprocessors can include independent thread scheduling capability to enable finer synchronization and cooperation between parallel threads. Streaming microprocessors can include a combined L1 data cache and shared memory unit to improve performance while simplifying programming.

[0080] The GPU(s) 908 may include high-bandwidth memory (HBM) and / or a 16 GB HBM2 memory subsystem to provide, in some examples, a peak memory bandwidth of approximately 900 GB / second. In some examples, synchronous graphics random access memory (SGRAM), such as double data rate type five synchronous graphics random access memory (GDDR5), may be used in addition to or alternatively to the HBM memory.

[0081] The GPU(s) 908 may include unified memory technology that includes access counters to enable more accurate migration of memory pages to the processor that accesses them most frequently, thereby improving efficiency for shared memory areas between processors. In some examples, address translation service (ATS) support may be used to enable the GPU(s) 908 to directly access the page tables of the CPU(s) 906. In such examples, when the memory management unit (MMU) of the GPU(s) 908 experiences a fault, an address translation request may be transmitted to the CPU(s) 906. In response, the CPU(s) 906 may look up the virtual-to-physical mapping for the address in their page tables and transmit the translation back to the GPU(s) 908.Thus, the unified memory technology may enable a single unified virtual address space for memory of both the CPU(s) 906 and the GPU(s) 908, thereby simplifying the programming and porting of GPU(s) 908 applications to the GPU(s) 908.

[0082] Additionally, the GPU(s) 908 may include an access counter that can track the frequency of the GPU(s) 908 accessing memory of other processors. The access counter can help ensure that memory pages are moved to the physical memory of the processor that accesses the pages most frequently.

[0083] The SoC(s) 904 may include any number of cache(s) 912, including those described herein. For example, the cache(s) 912 may include an L3 cache available to both the CPU(s) 906 and the GPU(s) 908 (e.g., connected to both the CPU(s) 906 and the GPU(s) 908). The cache(s) 912 may include a write-back cache that can track line states, such as by using a cache coherence protocol (e.g., MEI, MESI, MSI, etc.). The L3 cache may include 4 MB or more, depending on the embodiment, although smaller cache sizes may be used.

[0084] The SoC(s) 904 may include an arithmetic logic unit(s) (ALU(s)) that may be utilized in performing processing related to any of the vehicle's 900 variety of tasks or operations, such as processing DNNs. Additionally, the SoC(s) 904 may include a floating-point unit(s) (FPU(s))—or other types of mathematical coprocessors or numerical coprocessors—for performing mathematical operations within the system. For example, the SoC(s) 104 may include one or more FPUs integrated as execution units within a CPU(s) 906 and / or GPU(s) 908.

[0085] The SoC(s) 904 may include one or more accelerators 914 (e.g., hardware accelerators, software accelerators, or a combination thereof). For example, the SoC(s) 904 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., 4 MB SRAM) may enable the hardware acceleration cluster to accelerate neural networks and other computations. The hardware acceleration cluster may be used to supplement the GPU(s) 908 and offload some of the tasks of the GPU(s) 908 (e.g., to free up more cycles of the GPU(s) 908 to perform other tasks). As an example, the accelerator(s) 914 may be used for targeted workloads (e.g., perception, convolutional neural networks (CNNs), etc.) that are stable enough to be accelerable.The term “CNN” as used herein can include all types of CNNs, including region-based or regional convolutional neural networks (RCNNs) and fast RCNNs (e.g., as used for object detection).

[0086] The accelerator(s) 914 (e.g., the hardware acceleration cluster) may include one or more deep learning accelerators (DLAs). The DLA(s) may include one or more tensor processing units (TPUs), which may be configured to provide an additional tens of trillion operations per second for deep learning applications and inference. The TPUs may be accelerators configured and optimized to perform image processing functions (e.g., for CNNs, RCNNs, etc.). The DLA(s) may be further optimized for a specific set of neural network types and floating-point operations, as well as inference. The design of the DLA(s) may provide more performance per millimeter than a general-purpose GPU and significantly exceeds the performance of a CPU.The TPU(s) can perform several functions, including a single-instance convolution function that supports, for example, INT8, INT16, and FP16 data types for both features and weights, as well as post-processing functions.

[0087] The DLA(s) may quickly and efficiently execute neural networks, particularly CNNs, on processed or unprocessed data for any of a variety of functions, including, for example and without limitation: a CNN for object identification and detection using data from camera sensors; a CNN for distance estimation using data from camera sensors; a CNN for emergency vehicle detection and identification using data from microphones; a CNN for facial recognition and vehicle owner identification using data from camera sensors; and / or a CNN for safety and / or security-related events.

[0088] The DLA(s) can perform any function of the GPU(s) 908, and by using an inference accelerator, a designer can, for example, target either the DLA(s) or the GPU(s) 908 for any function. For example, the designer can focus the processing of CNNs and floating-point operations on the DLA(s) and leave other functions to the GPU(s) 908 and / or the other accelerator(s) 914.

[0089] The accelerator(s) 914 (e.g., the hardware acceleration cluster) may include one or more programmable vision accelerators (PVAs), which may alternatively be referred to herein as a computer vision accelerator. The PVA(s) 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(s) may provide a balance between performance and flexibility. For example, each PVA(s) may include, for example and without limitation, any number of reduced instruction set computing (RISC) cores, direct memory access (DMA), and / or any number of vector processors.

[0090] The RISC cores may interact with image sensors (e.g., the image sensors of any of the cameras described herein), image signal processor(s), and / or the like. Each of the RISC cores may include any amount of memory. The RISC cores may use any of a number of protocols, depending on the embodiment. In some examples, the RISC cores may execute a real-time operating system (RTOS). The RISC cores may be implemented using one or more integrated circuit devices, application-specific integrated circuits (ASICs), and / or memory devices. For example, the RISC cores may include an instruction cache and / or tightly coupled RAM.

[0091] The DMA may enable components of the PVA(s) to access system memory independently of the CPU(s) 906. The DMA may support any number of features used to provide optimization to the PVA, including, but not limited to, supporting multi-dimensional addressing and / or circular addressing. In some examples, the DMA may support up to six or more dimensions of addressing, which may include block width, block height, block depth, horizontal block increments, vertical block increments, and / or depth increments.

[0092] The vector processors may be programmable processors that can be designed to efficiently and flexibly execute programming for computer vision algorithms and provide signal processing capabilities. In some examples, the PVA may include a PVA core and two vector processing subsystem partitions. The PVA core may include a processor subsystem, DMA engine(s) (e.g., two DMA engines), and / or other peripherals. The vector processing subsystem may operate as the primary processing engine of the PVA and may include a vector processing unit (VPU), an instruction cache, and / or vector memory (e.g., VMEM). A VPU core may include a digital signal processor, such as, for example, a single instruction multiple data (SIMD) very long instruction word (VLIW) digital signal processor. The combination of SIMD and VLIW can improve throughput and speed.

[0093] Each of the vector processors may include an instruction cache and may be coupled to 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, the plurality of 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 concurrently execute different computer vision algorithms on the same image or even execute different algorithms on consecutive images or portions of an image.Among other things, any number of PVAs may be included in the hardware acceleration cluster, and any number of vector processors may be included in each of the PVAs. Additionally, the PVA(s) may include additional error correction code (ECC) memory to improve overall system security.

[0094] The accelerator(s) 914 (e.g., the hardware acceleration cluster) may include an on-chip and SRAM computer vision network for providing high-bandwidth, low-latency SRAM to the accelerator(s) 914. In some examples, the on-chip memory may include at least 4 MB of SRAM, consisting of, for example, and without limitation, eight field-configurable memory blocks accessible 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 the DLA may access the memory via a backbone that provides high-speed access to the memory to the PVA and the DLA. The backbone may include an on-chip computer vision network connecting the PVA and the DLA to the memory (e.g.,using the APB).

[0095] The on-chip computer vision network may include an interface that determines that both the PVA and the DLA are ready and providing valid signals before transmitting any control signal / address / data. Such an interface may provide separate phases and separate channels for transmitting control signals / addresses / data, as well as burst-type communications for continuous data transmission. This type of interface may conform to ISO 26262 or IEC 61508, although other standards and protocols may be used.

[0096] In some examples, the SoC(s) 904 may include a real-time ray tracing hardware accelerator, such as described in U.S. Patent Application No. 16 / 101,232, filed August 10, 2018. The real-time ray tracing hardware accelerator may be used to quickly and efficiently determine the positions and extents of objects (e.g., within a world model) to generate real-time visualization simulations for radar signal interpretation, for sound propagation synthesis and / or analysis, for simulation of sonar systems, for general wave propagation simulation, for comparison with LiDAR data for localization and / or other functions, and / or for other uses. In some embodiments, one or more tree traversal units (TTUs) may be used to perform one or more ray tracing-related operations.

[0097] The accelerator(s) 914 (e.g., the hardware accelerator cluster) have a wide range of uses for autonomous driving. The PVA can be a programmable vision 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 with low power and low latency. In other words, the PVA performs well on semi-dense or dense regular computation, even on small datasets that require predictable runtimes with low latency and low power. Thus, in the context of autonomous vehicle platforms, the PVAs are designed to execute classical computer vision algorithms because they are efficient at object detection and operate on integer mathematics.

[0098] For example, according to one embodiment of the technology, the PVA is used to perform computer stereo vision. A semi-global matching-based algorithm may be used in some examples, although this is not intended to be limiting. Many Level 3-5 autonomous driving applications require on-the-fly motion estimation / stereo matching (e.g., structure from motion, pedestrian detection, lane detection, etc.). The PVA can perform a computer stereo vision function on inputs from two monocular cameras.

[0099] In some examples, the PVA can be used to perform dense optical flow. According to the processing, raw radar data is processed (e.g., using a 4D fast Fourier transform) to provide processed radar. In other examples, the PVA is used for runtime-depth processing, for example, by processing raw runtime data to provide processed runtime data.

[0100] The DLA can be used to run any type of network to improve 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 allows the system to make further decisions about which detections should be considered true positives rather than false positives. For example, the system can set a confidence threshold and consider only those detections that exceed the threshold as true positives. In an automatic emergency braking (AEB) system, false positives would cause the vehicle to automatically perform emergency braking, which is obviously undesirable.Therefore, only the most confident detections should be considered as triggers for AEB. The DLA can execute a neural network to reduce the confidence value. The neural network can take as its input at least a subset of parameters, such as bounding box dimensions, ground plane estimation obtained (e.g., from another subsystem), output from an inertial measurement unit (IMU) sensor 966 correlated with the orientation of the vehicle 900, distance, 3D position estimates of the object obtained by the neural network, and / or other sensors (e.g., LiDAR sensor(s) 964 or RADAR sensor(s) 960), among others.

[0101] The SoC(s) 904 may include data storage 916 (e.g., memory). The data storage 916 may be on-chip memory of the SoC(s) 904, which may store neural networks to be executed on the GPU and / or the DLA. In some examples, the data storage 916 may be large enough in capacity to store multiple neural network instances for redundancy and security. The data storage 912 may include L2 or L3 cache(s). Reference to the data storage 916 may include reference to the memory associated with the PVA, the DLA, and / or other accelerator(s) 914, as described herein.

[0102] The SoC(s) 904 may include one or more processor(s) 910 (e.g., embedded processors). The processor(s) 910 may include a boot and power management processor, which may be a dedicated processor and subsystem to handle boot power and management functions and associated security enforcement. The boot and power management processor may be part of the boot sequence of the SoC(s) 904 and may provide runtime power management services. The boot power and management processor may provide clock and voltage programming, support for system low-power state transitions, management of thermal and temperature sensors of the SoC(s) 904, and / or management of the power states of the SoC(s) 904.Each temperature sensor may be implemented as a ring oscillator whose output frequency is proportional to temperature, and the SoC(s) 904 may use the ring oscillators to detect temperatures of the CPU(s) 906, GPU(s) 908, and / or accelerator 914. If temperatures are determined to exceed a threshold, the boot and power management processor may enter a temperature fault routine and place the SoC(s) 904 into a lower power state and / or place the vehicle 900 into a chauffeur-safe stop mode (e.g., bring the vehicle 900 to a safe stop).

[0103] The processor(s) 910 may further include a set of embedded processors that may serve as an audio processing engine. The audio processing engine may be an audio subsystem that enables full hardware support for multi-channel audio across multiple interfaces and a wide and flexible range of audio I / O interfaces. In some examples, the audio processing engine is a dedicated processor core with a digital signal processor with dedicated RAM.

[0104] The processor(s) 910 may further include an always-on processor engine that may provide necessary hardware features to support low-power sensor management and wake-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 control peripherals, and routing logic.

[0105] The processor(s) 910 may further include a security cluster engine, which includes a dedicated processor subsystem to handle security management for automotive applications. The security cluster engine may include two or more processor cores, tightly coupled RAM, supporting peripherals (e.g., timers, an interrupt controller, etc.), and / or routing logic. In a security mode, the two or more cores may operate in a lockstep mode, functioning as a single core with comparison logic to detect differences between their operations.

[0106] The processor(s) 910 may further include a real-time camera engine, which may include a dedicated processor subsystem for handling real-time camera management.

[0107] The processor(s) 910 may further include a high dynamic range signal processor, which may include an image signal processor that is a hardware engine that is part of the camera processing pipeline.

[0108] The processor(s) 910 may include a video image compositor, which may be a processing block (e.g., implemented on a microprocessor) that implements video post-processing functions required by a video playback application to generate the final image for the playback window. The video image compositor may perform lens distortion correction on the wide-view camera(s) 970, the surround camera(s) 974, and / or on in-cabin monitoring camera sensors. The in-cabin monitoring camera sensor is preferably monitored by a neural network executing on another instance of the Advanced SoC configured to identify in-cabin events and respond accordingly.An in-cabin system can perform lip reading to activate cellular service and place a phone call, dictate emails, change the vehicle's destination, activate or change the vehicle's infotainment system and settings, or provide voice-activated web browsing. Certain functions are available to the driver only when the vehicle is operating in autonomous mode and are disabled otherwise.

[0109] The video image compositor may incorporate enhanced temporal noise reduction for both spatial and temporal noise reduction. For example, when motion occurs in a video, the noise reduction appropriately weights spatial information, thereby reducing the weight of information provided by neighboring frames. When an image or a portion of an image contains no motion, the temporal noise reduction performed by the video image compositor may use information from the previous frame to reduce noise in the current frame.

[0110] The video image compositor may also be configured to perform stereo equalization on input stereo lens frames. The video image compositor may further be used for user interface composition when the operating system desktop is in use and the GPU(s) 908 are not required to continuously render new surfaces. Even when the GPU(s) 908 are turned on and active to perform 3D rendering, the video image compositor may be used to offload the GPU(s) 908 to improve performance and responsiveness.

[0111] The SoC(s) 904 may further include a Mobile Industrial Processor Interface (MIPI) serial camera interface for receiving video and input from cameras, a high-speed interface, and / or a video input block that may be used for camera and related pixel input functions. The SoC(s) 904 may further include an input / output controller(s) that may be controlled by software and may be used to receive I / O signals that are not committed to a specific role.

[0112] The SoC(s) 904 may further include a wide range of peripheral interfaces to enable communication with peripherals, audio codecs, power management, and / or other devices. The SoC(s) 904 may be used to process data from cameras (e.g., connected via Gigabit Multimedia Serial Link and Ethernet), sensors (e.g., LiDAR sensor(s) 964, RADAR sensor(s) 960, etc., which may be connected via Ethernet), data from the bus 902 (e.g., speed of the vehicle 900, steering wheel position, etc.), and data from the GNSS sensor(s) 958 (e.g., connected via Ethernet or CAN bus). The SoC(s) 904 may further include dedicated high-performance mass storage controllers, which may include their own DMA engines and which may be used to free the CPU(s) 906 from routine data management tasks.

[0113] The SoC(s) 904 may be an end-to-end platform with a flexible architecture spanning automation levels 3-5, providing a comprehensive functional safety architecture that leverages and efficiently uses computer vision and ADAS techniques for diversity and redundancy, and providing a platform for a flexible, reliable driving software stack along with deep learning tools. The SoC(s) 904 may be faster, more reliable, and even more power- and space-efficient than conventional systems. For example, the accelerator(s) 914, when combined with the CPU(s) 906, the GPU(s) 908, and the memory(s) 916, may provide a fast, efficient platform for autonomous vehicles at levels 3-5.

[0114] 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 perform a wide variety of processing algorithms on a wide variety of visual data. However, CPUs are often unable to meet the performance requirements of many computer vision applications, such as those related to execution time and power consumption. In particular, many CPUs are unable to execute complex object detection algorithms in real time, which is a requirement of in-vehicle ADAS applications and a requirement for practical Level 3-5 autonomous vehicles.

[0115] Unlike conventional systems, the technology described herein, by providing a CPU complex, a GPU complex, and a hardware acceleration cluster, enables multiple neural networks to be executed simultaneously and / or sequentially, and for the results to be combined to enable Level 3-5 autonomous driving functionality. For example, a CNN executing on the DLA or dGPU (e.g., GPU(s) 920) may include text and word recognition, allowing the supercomputer to read and understand traffic signs, including signs for which the neural network has not been specifically trained. The DLA may further include a neural network capable of identifying, interpreting, and providing a semantic understanding of the sign, and passing this semantic understanding to the path planning modules executing on the CPU complex.

[0116] As another example, multiple neural networks can be executed simultaneously, as required for Level 3, 4, or 5 driving. For example, a warning sign consisting of "Warning: Flashing lights indicate icy conditions," along with an electrical light, can be interpreted independently or jointly by multiple neural networks. The sign itself can be identified as a traffic sign by a first deployed neural network (e.g., a trained neural network), and the text "Flashing lights indicate icy conditions" can be interpreted by a second deployed neural network, which informs the vehicle's path-planning software (preferably running on the CPU complex) that if flashing lights are detected, icy conditions exist.The flashing light can be identified by running a third deployed neural network over multiple frames, informing the vehicle's path planning software of the presence (or absence) of flashing lights. All three neural networks can be executed simultaneously, such as within the DLA and / or on the GPU(s) 908.

[0117] In some examples, a CNN for facial recognition and vehicle owner identification may use data from camera sensors to identify the presence of an authorized driver and / or owner of vehicle 900. The always-on sensor processing engine may be used to unlock the vehicle when the owner approaches the driver's door and turn on the lights, and in secure mode, disarm the vehicle when the owner exits the vehicle. In this way, the SoC(s) 904 provide security against theft and / or carjacking.

[0118] In another example, a CNN for emergency vehicle detection and identification may use data from microphones 996 to detect and identify emergency vehicle sirens. Unlike conventional systems that use general classifiers to detect sirens and manually extract features, the SoC(s) 904 use the CNN to classify ambient and urban noise, as well as visual data. In a preferred embodiment, the CNN running on the DLA is trained to identify the relative closing speed of the emergency vehicle (e.g., using the Doppler effect). The CNN may also be trained to identify emergency vehicles specific to the local area in which the vehicle is operating, as identified by the GNSS sensor(s) 958.For example, if the CNN is operating in Europe, it will attempt to detect European sirens, and if the CNN is operating in the United States, it will attempt to identify only North American sirens. Once an emergency vehicle is detected, a control program can be used to execute an emergency vehicle safety routine, slow the vehicle, pull it to the side of the road, park the vehicle, and / or idle the vehicle using ultrasonic sensors 962 until the emergency vehicle(s) pass by.

[0119] The vehicle may include a CPU(s) 918 (e.g., discrete CPU(s) or dCPU(s)) that may be coupled to the SoC(s) 904 via a high-speed connection (e.g., PCIe). The CPU(s) 918 may include, for example, an x86 processor. The CPU(s) 918 may be used to perform any of a variety of functions, including, for example, arbitrating potentially inconsistent results between ADAS sensors and the SoC(s) 904 and / or monitoring the status and health of the controller(s) 936 and / or the infotainment SoC 930.

[0120] The vehicle 900 may include a GPU(s) 920 (e.g., discrete GPU(s) or dGPU(s)) that may be coupled to the SoC(s) 904 via a high-speed interconnect (e.g., NVIDIA's NVLINK). The GPU(s) 920 may provide additional artificial intelligence functionality, such as by executing redundant and / or different neural networks, and may be used to train and / or update neural networks based on inputs (e.g., sensor data) from sensors of the vehicle 900.

[0121] The vehicle 900 may further include the network interface 924, which may include one or more wireless antennas 926 (e.g., one or more wireless antennas for various communication protocols, such as a cellular antenna, a Bluetooth antenna, etc.). The network interface 924 may be used to enable wireless connectivity over the internet to the cloud (e.g., to the server(s) 978 and / or other network devices), to other vehicles, and / or to computing devices (e.g., passenger client devices). To communicate with other vehicles, a direct connection may be established between the two vehicles and / or an indirect connection may be established (e.g., via networks and over the internet). Direct connections may be provided using a vehicle-to-vehicle communication link.The vehicle-to-vehicle communication link may provide information to the vehicle 900 about vehicles in the vicinity of the vehicle 900 (e.g., vehicles in front of, to the side of, and / or behind the vehicle 900). This functionality may be part of a cooperative adaptive cruise control functionality of the vehicle 900.

[0122] The network interface 924 may include an SoC that provides modulation and demodulation functionality and enables the controller(s) 936 to communicate over wireless networks. The network interface 924 may include a radio frequency front-end for upconverting from baseband to radio frequency and downconverting from radio frequency to baseband. The frequency conversions may be performed by well-known processes and / or may be performed using superheterodyne processes. In some examples, the radio frequency front-end functionality may be provided by a separate chip. The network interface may include wireless functionality for communicating over LTE, WCDMA, UMTS, GSM, CDMA2000, Bluetooth, Bluetooth LE, Wi-Fi, Z-Wave, ZigBee, LoRaWAN, and / or other wireless protocols.

[0123] The vehicle 900 may further include one or more data stores 928, which may include off-chip memory (e.g., outside the SoC(s) 904). The data stores 928 may include one or more memory elements, including RAM, SRAM, DRAM, VRAM, flash, hard drives, and / or other components and / or devices capable of storing at least one bit of data.

[0124] The vehicle 900 may further include one or more GNSS sensors 958. The GNSS sensor(s) 958 (e.g., GPS, assisted GPS sensors, differential GPS (DGPS) sensors, etc.) may assist in mapping, sensing, occupancy grid generation, and / or path planning functions. Any number of GNSS sensors 958 may be used, including, for example, and without limitation, a GPS using a USB connector with an Ethernet-to-serial (RS-232) bridge.

[0125] The vehicle 900 may further include one or more RADAR sensors 960. The RADAR sensor(s) 960 may be used by the vehicle 900 for long-range vehicle detection, even in darkness and / or severe weather conditions. RADAR functional safety levels may be ASIL B. The RADAR sensor(s) 960 may use the CAN and / or bus 902 (e.g., to transmit data generated by the RADAR sensor(s) 960) for control and, in some examples, access object tracking data with access to Ethernet for raw data. A wide variety of RADAR sensor types may be used. For example, and without limitation, the RADAR sensor(s) 960 may be suitable for front, rear, and side RADAR use. In some examples, one or more Pulse Doppler RADAR sensors are used.

[0126] The RADAR sensor(s) 960 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, etc. In some examples, long-range RADAR may be used for adaptive cruise control functionality. Long-range RADAR systems may provide a wide field of view realized by two or more independent scans, such as within a 250 m range. The RADAR sensor(s) 960 may help distinguish between static and moving objects and may be used by ADAS systems for emergency braking assistance and forward collision warning. Long-range RADAR sensors may include monostatic multimodal RADAR with multiple (e.g., six or more) fixed RADAR antennas and a high-speed CAN and FlexRay interface.In a six-antenna example, the middle four antennas can create a focused beam pattern designed to record the surroundings of vehicle 900 at higher speeds with minimal interference from traffic in adjacent lanes. The other two antennas can expand the field of view, making it possible to quickly detect vehicles entering or exiting the lane of vehicle 900.

[0127] Medium-range radar systems, for example, can have a range of up to 960 m (front) or 80 m (rear) and a field of view of up to 42 degrees (front) or 950 degrees (rear). Short-range radar systems can, without limitation, include radar sensors designed to be installed at both ends of the rear bumper. When installed at both ends of the rear bumper, such radar sensor systems can generate two beams that continuously monitor the blind spot area to the rear and side of the vehicle.

[0128] Short-range radar systems can be used in an ADAS system for blind spot detection and / or lane change assistance.

[0129] The vehicle 900 may further include one or more ultrasonic sensors 962. The ultrasonic sensor(s) 962, which may be positioned at the front, rear, and / or sides of the vehicle 900, may be used for parking assistance and / or for creating and updating an occupancy grid. A wide variety of ultrasonic sensors 962 may be used, and different ultrasonic sensors 962 may be used for different detection ranges (e.g., 2.5 m, 4 m). The ultrasonic sensor(s) 962 may operate at functional safety levels of ASIL B.

[0130] The vehicle 900 may include one or more LiDAR sensors 964. The LiDAR sensor(s) 964 may be used for object and pedestrian detection, emergency braking, collision avoidance, and / or other functions. The LiDAR sensor(s) 964 may be at functional safety level ASIL B. In some examples, the vehicle 900 may include multiple LiDAR sensors 964 (e.g., two, four, six, etc.) that may use Ethernet (e.g., to provide data to a Gigabit Ethernet switch).

[0131] In some examples, the LiDAR sensor(s) 964 may be capable of providing a list of objects and their distances for a 360-degree field of view. The commercially available LiDAR sensor(s) 964 may have an advertised range of approximately 900 m with an accuracy of 2 cm-3 cm and with support for, for example, a 900 Mbps Ethernet connection. In some examples, one or more non-prominent LiDAR sensors 964 may be used. In such examples, the LiDAR sensor(s) 964 may be implemented as a small device that may be embedded in the front, rear, sides, and / or corners of the vehicle 900. In such examples, the LiDAR sensor(s) 964 may provide up to a 120-degree horizontal field of view and a 35-degree vertical field of view with a range of 200 m even for low-reflectivity objects.The front-mounted LiDAR sensor(s) 964 may be configured for a horizontal field of view between 45 degrees and 135 degrees.

[0132] In some examples, LiDAR technologies such as 3D Flash LiDAR may also be used. 3D Flash LiDAR uses a laser flash as a transmission source to illuminate the vehicle's surroundings up to approximately 200 m. A Flash LiDAR unit includes a receiver that records the laser pulse travel time and reflected light at each pixel, which in turn corresponds to the range from the vehicle to the objects. Flash LiDAR can enable highly accurate and distortion-free images of the surroundings to be created with each laser flash. In some examples, four Flash LiDAR sensors may be deployed, one on each side of the vehicle. Available 3D Flash LiDAR systems include a solid-state 3D staring array LiDAR camera with no moving parts other than a fan (e.g., a non-scanning LiDAR device).The flash LiDAR device can use a 5-nanosecond Class I (eye-safe) laser pulse per frame and can capture the reflected laser light as 3D area point clouds 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(s) 964 can be less susceptible to motion blur, vibration, and / or shock.

[0133] The vehicle may further include one or more IMU sensors 966. In some examples, the IMU sensor(s) 966 may be positioned at a center of the rear axle of the vehicle 900. The IMU sensor(s) 966 may include, for example, and without limitation, an accelerometer, a magnetometer, a gyroscope, a magnetic compass, and / or other types of sensors. In some examples, such as in six-axis applications, the IMU sensor(s) 966 may include accelerometers and gyroscopes, while in nine-axis applications, the IMU sensor(s) 966 may include accelerometers, gyroscopes, and magnetometers.

[0134] In some embodiments, the IMU sensor(s) 966 may be implemented as a miniature, high-performance GPS-based inertial navigation system (GPS / INS) that combines microelectromechanical system (MEMS) inertial sensors, a high-sensitivity GPS receiver, and advanced Kalman filter algorithms to provide estimates of position, velocity, and attitude. Thus, in some examples, the IMU sensor(s) 966 may enable the vehicle 900 to estimate heading without requiring input from a magnetic sensor by directly observing changes in velocity from the GPS and correlating them with the IMU sensor(s) 966. In some examples, the IMU sensor(s) 966 and the GNSS sensor(s) 958 may be combined into a single integrated unit.

[0135] The vehicle may include one or more microphones 996 located in and / or around the vehicle 900. The microphone(s) 996 may be used, among other things, for emergency vehicle detection and identification.

[0136] The vehicle may further include any number of camera types, including stereo camera(s) 968, wide-view camera(s) 970, infrared camera(s) 972, surround camera(s) 974, long-range and / or medium-range camera(s) 998, and / or other camera types. The cameras may be used to capture image data around an entire periphery of the vehicle 900. The camera types used depend on the embodiments and requirements for the vehicle 900, and any combination of camera types may be used to provide the necessary coverage around the vehicle 900. Additionally, 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. The cameras may support, by way of example and without limitation, Gigabit Multimedia Serial Link (GMSL) and / or Gigabit Ethernet.Each of the camera(s) is referred to herein in relation to . Fig. 9A and Fig. 9B is described in more detail.

[0137] The vehicle 900 may further include one or more vibration sensors 942. The vibration sensor(s) 942 may measure vibrations from components of the vehicle, such as the axle(s). Changes in vibration may, for example, indicate a change in the road surfaces. In another example, when two or more vibration sensors 942 are used, the differences between the vibrations may be used to determine friction or slippage of the road surface (e.g., if the difference in vibration is between a driven axle and a free-spinning axle).

[0138] The vehicle 900 may include an ADAS system 938. The ADAS system 938 may include an SoC in some examples. The ADAS system 938 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 assistance (LKA), blind spot warning (BSW), rear cross-traffic alert (RCTW), collision warning systems (CWS), lane centering (LC), and / or other features and functionality.

[0139] The ACC systems may use one or more radar sensors 960, one or more lidar sensors 964, and / or one or more cameras. The ACC systems may include longitudinal ACC and / or lateral ACC. Longitudinal ACC monitors and controls the distance to the vehicle immediately ahead of the vehicle 900 and automatically adjusts the vehicle speed to maintain a safe distance from preceding vehicles. Lateral ACC performs follow-through and advises the vehicle 900 to change lanes if necessary. Lateral ACC refers to other ADAS applications, such as LCA and CWS.

[0140] CACC utilizes information from other vehicles, which may be received via the network interface 924 and / or the wireless antenna(s) 926 from other vehicles over a wireless connection or indirectly via a network connection (e.g., over the Internet). Direct connections may be provided through a vehicle-to-vehicle (V2V) communication link, while indirect connections may be infrastructure-to-vehicle (I2V) communication links. In general, the V2V communication concept provides information about the immediately preceding vehicles (e.g., vehicles immediately in front of and in the same lane as vehicle 900), while the I2V communication concept provides information about traffic further ahead. CACC systems may incorporate one or both I2V and V2V information sources.Given the information from the vehicles ahead of vehicle 900, CACC can be more reliable and has the potential to improve traffic flow smoothness and reduce congestion on the road.

[0141] FCW systems are designed to warn the driver of a hazard so the driver can take corrective action. FCW systems utilize a forward-facing camera and / or one or more radar sensors 960 coupled to a dedicated processor, DSP, FPGA, and / or ASIC, which is / are electrically coupled to driver feedback, such as a display, speaker, and / or vibration component. FCW systems can provide a warning, such as a sound, a visual warning, vibration, and / or a rapid braking pulse.

[0142] AEB systems detect an impending forward collision with another vehicle or object and can automatically apply the brakes if the driver does not take corrective action within a specified time or distance parameter. AEB systems can utilize one or more forward-facing cameras and / or one or more radar sensors 960 coupled with 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. If the driver does not take corrective action, the AEB system can automatically apply the brakes to prevent or at least mitigate the effects of the predicted collision. AEB systems can incorporate techniques such as dynamic brake support and / or imminent collision braking.

[0143] LDW systems provide visual, audible, and / or tactile warnings, such as steering wheel or seat vibrations, to alert the driver when the vehicle crosses lane markings. An LDW system will not activate if the driver indicates an intended lane departure by activating a turn signal. LDW systems may utilize forward-facing cameras coupled with a dedicated processor, DSP, FPGA, and / or ASIC, which is / are electrically coupled to driver feedback, such as a display, speaker, and / or vibration component.

[0144] LKA systems are a variation of LDW systems. LKA systems provide steering input or braking to correct the vehicle 900 when the vehicle 900 begins to drift out of its lane.

[0145] BSW systems detect and warn the driver of vehicles in an automobile's blind spot. BSW systems may provide a visual, audible, and / or tactile warning to indicate that merging or changing lanes is unsafe. The system may provide an additional warning when the driver uses a turn signal. BSW systems may utilize one or more rear-facing cameras and / or one or more radar sensors 960 coupled to a dedicated processor, DSP, FPGA, and / or ASIC electrically coupled to driver feedback, such as a display, speaker, and / or vibration component.

[0146] RCTW systems may provide visual, audible, and / or tactile notification when an object is detected outside the rear camera range when the vehicle 900 is reversing. Some RCTW systems include AEB to ensure the vehicle brakes are applied to avoid a collision. RCTW systems may utilize one or more rear-facing RADAR sensors 960 coupled to a dedicated processor, DSP, FPGA, and / or ASIC electrically coupled to driver feedback, such as a display, speaker, and / or vibration component.

[0147] Conventional ADAS systems can be prone to false positives, which can be annoying and distracting for a driver, but are typically not catastrophic because the ADAS systems alert the driver and allow the driver to decide whether a safety condition actually exists and act accordingly. However, in an autonomous vehicle 900, even in the case of conflicting results, the vehicle 900 must decide whether to honor the result from a primary computer or a secondary computer (e.g., a first controller 936 or a second controller 936). For example, in some embodiments, the ADAS system 938 may be a backup and / or secondary computer for providing perception information to a backup computer rationality module.The backup computer rationality monitor can run redundant, diverse software on hardware components to detect errors in perception and dynamic driving tasks. Outputs from the ADAS system 938 can be provided to a monitoring MCU. If outputs from the primary computer and the secondary computer conflict, the monitoring MCU must determine how to reconcile the conflict to ensure safe operation.

[0148] In some examples, the primary computer may be configured to provide the monitoring MCU with a confidence score indicating the primary computer's confidence in the selected outcome. If the confidence score exceeds a threshold, the monitoring MCU may follow the primary computer's instruction regardless of whether the secondary computer provides a conflicting or inconsistent outcome. If the confidence score does not meet the threshold and if the primary and secondary computers indicate different outcomes (e.g., conflict), the monitoring MCU may arbitrate between the computers to determine the appropriate outcome.

[0149] The monitoring MCU may be configured to run a neural network(s) trained and configured to determine, based on outputs from the primary computer and the secondary computer, conditions under which the secondary computer provides false alarms. Thus, the neural network(s) in the monitoring MCU can learn when the output of the secondary computer can be trusted and when it cannot. For example, if the secondary computer is a radar-based FCW system, a neural network(s) in the monitoring MCU can learn when the FCW system identifies metallic objects that are not actually hazards, such as a drainage grate or manhole cover, which triggers an alarm.Similarly, if the secondary computer is a camera-based LDW system, a neural network in the monitoring MCU can learn to override the LDW when cyclists or pedestrians are present and lane departure is actually the safest maneuver. In embodiments including neural network(s) executing on the monitoring MCU, the monitoring MCU can include at least one of a DLA or a GPU suitable for executing the neural network(s) with associated memory. In preferred embodiments, the monitoring MCU can include and / or be included as a component of the SoC(s) 904.

[0150] In other examples, the ADAS system 938 may include a secondary computer that performs ADAS functionality using conventional computer vision rules. Thus, the secondary computer may use classic computer vision rules (if-then), and the presence of a neural network(s) in the monitoring MCU may improve reliability, safety, and performance. For example, the diverse implementation and intended non-identity makes the overall system more fault-tolerant, particularly toward errors caused by software (or software-hardware interface) functionality.For example, if there is a software bug or error in the software running on the primary computer, and the non-identical software code running on the secondary computer provides the same overall result, the monitoring MCU can have greater confidence that the overall result is correct, and the software or hardware error on the primary computer does not cause a material defect.

[0151] In some examples, the output of the ADAS system 938 may be fed into the perception block of the primary computer and / or the dynamic driving task block of the primary computer. For example, if the ADAS system 938 indicates a forward collision warning due to an object immediately ahead, the perception block may use this information in identifying objects. In other examples, the secondary computer may have its own neural network trained, thus reducing the risk of false positives, as described herein.

[0152] The vehicle 900 may further include the infotainment SoC 930 (e.g., an in-vehicle infotainment (IVI) system). 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 930 may include a combination of hardware and software that can be used to provide the vehicle 900 with audio (e.g., music, a personal digital assistant, navigation directions, news, radio, etc.), video (e.g., television, movies, streaming, etc.), phone (e.g., hands-free calling), network connectivity (e.g., LTE, Wi-Fi, etc.), and / or information services (e.g., navigation systems, backup assist, a radio data system, vehicle-related information such as fuel level, total distance traveled, brake fuel level, oil level, door open / close, air filter information, etc.).For example, the infotainment SoC 930 may include radios, record players, navigation systems, video players, USB and Bluetooth connectivity, car stereos, in-vehicle entertainment, Wi-Fi, steering wheel audio controls, hands-free voice control, a front-end display (HUD), a human-machine interface (HMI) display 934, 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 930 may further be used to provide information (e.g., visual and / or audible) to a user(s) of the vehicle, such as information from the ADAS system 938, autonomous driving information such as planned vehicle maneuvers, trajectories, environmental information (e.g., intersection information, vehicle information, road information, etc.), and / or other information.

[0153] The infotainment SoC 930 may include GPU functionality. The infotainment SoC 930 may communicate with other devices, systems, and / or components of the vehicle 900 via the bus 902 (e.g., CAN bus, Ethernet, etc.). In some examples, the infotainment SoC 930 may be coupled to a supervisory MCU so that the GPU of the infotainment system can perform some self-driving functions in the event that the primary controller(s) 936 (e.g., the primary and / or backup computer of the vehicle 900) fails. In such an example, the infotainment SoC 930 may place the vehicle 900 into a chauffeur-in-safe-stop mode, as described herein.

[0154] The vehicle 900 may further include an instrument cluster 932 (e.g., a digital instrument panel, an electronic instrument cluster, a digital dashboard, etc.). The instrument cluster 932 may include a controller and / or a supercomputer (e.g., a discrete controller or a supercomputer). The instrument cluster 932 may include a set of instruments, such as a speedometer, fuel level, oil pressure, tachometer, odometer, turn signals, gear shift position indicator, seat belt warning light(s), parking brake warning light(s), engine failure light(s), airbag (SRS) system information, lighting controls, safety system controls, navigation information, etc. In some examples, information may be displayed and / or shared between the infotainment SoC 930 and the instrument cluster 932. In other words, the instrument cluster 932 may be included as part of the infotainment SoC 930, or vice versa.

[0155] Fig. 9D is a system diagram for communication between one or more cloud-based servers and the example autonomous vehicle 900 of Fig. 9A, according to some embodiments of the present disclosure. The system 976 may include one or more servers 978, one or more networks 990, and vehicles, including the vehicle 900. The server(s) 978 may include a plurality of GPUs 984(A)-984(H) (collectively referred to herein as GPUs 984), PCIe switches 982(A)-982(H) (collectively referred to herein as PCIe switches 982), and / or CPUs 980(A)-980(B) (collectively referred to herein as CPUs 980). The GPUs 984, the CPUs 980, and the PCIe switches may be connected to high-speed interconnects, such as, without limitation, NVLink interfaces 988 developed by NVIDIA and / or PCIe interconnects 986. In some examples, the GPUs 984 are connected via NVLink and / or NVSwitch SoC, and the GPUs 984 and the PCIe switches 982 are connected via PCIe links.Although eight GPUs 984, two CPUs 980, and two PCIe switches are illustrated, this is not intended to be limiting. Depending on the embodiment, each of the server(s) 978 may include any number of GPUs 984, CPUs 980, and / or PCIe switches. For example, the server(s) 978 may include eight, sixteen, thirty-two, and / or more GPUs 984, respectively.

[0156] The server(s) 978 may receive, via the network(s) 990 and from the vehicles, image data representative of images depicting unexpected or changed road conditions, such as recently commenced roadwork. The server(s) 978 may transmit, via the network(s) 990 and to the vehicles, neural networks 992, updated neural networks 992, and / or map information 994, including information related to traffic and road conditions. The updates to the map information 994 may include updates to the HD map 922, such as information related to construction, potholes, detours, flooding, and / or other obstacles.In some examples, the neural networks 992, the updated neural networks 992, and / or the map information 994 may result from new training and / or experience represented in data received from any number of vehicles in the environment and / or based on training performed in a data center (e.g., using the server(s) 978 and / or other servers).

[0157] The server(s) 978 may be used to train machine learning models (e.g., neural networks) based on training data. The training data may be generated by the vehicles and / or may be generated in a simulation (e.g., using a gaming machine). In some examples, the training data is labeled (e.g., where the neural network benefits from supervised learning) and / or subjected to other preprocessing, while in other examples, the training data is unlabeled and / or preprocessed (e.g., where the neural network does not require supervised learning).Training may be performed according to any one or more classes of machine learning techniques, including, but not limited to, supervised training, semi-supervised training, unsupervised training, self-learning, reinforcement learning, federated learning, transfer learning, feature learning (including principal component and cluster analysis), multilinear subspace learning, manifold learning, representation learning (including free dictionary learning), rule-based machine learning, anomaly detection, and any variations or combinations thereof. Once the machine learning models are trained, the machine learning models may be used by the vehicles (e.g., transmitted to the vehicles via the network(s) 990) and / or the machine learning models may be used by the server(s) 978 to remotely monitor the vehicles.

[0158] In some examples, the server(s) 978 may receive data from the vehicles and apply the data to real-time neural networks for intelligent real-time inference building. The server(s) 978 may include deep learning supercomputers and / or dedicated AI computers powered by GPU(s) 984, such as DGX and DGX Station machines developed by NVIDIA. However, in some examples, the server(s) 978 may include deep learning infrastructure using only CPU-powered data centers.

[0159] The deep learning infrastructure of server(s) 978 may be capable of rapid, real-time inference formation and may use this capability to evaluate and verify the health of the processors, software, and / or associated hardware in vehicle 900. For example, the deep learning infrastructure may receive periodic updates from vehicle 900, such as a sequence of images and / or objects that vehicle 900 has positioned within that sequence of images (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 the objects and compare them to the objects identified by the vehicle 900, and if the results do not match and the infrastructure concludes that the AI ​​in the vehicle 900 is malfunctioning, the server(s) 978 may transmit a signal to the vehicle 900 instructing a fail-safe computer of the vehicle 900 to assume control, notify the passengers, and complete a safe parking maneuver.

[0160] For inference, the server(s) 978 may include the GPU(s) 984 and one or more programmable inference accelerators (e.g., NVIDIA's TensorRT). Combining servers with GPU performance and inference acceleration may enable real-time responsiveness. In other examples, such as when performance is less critical, servers powered by CPUs, FPGAs, and other processors may be used for inference. Example computing device

[0161] Fig. 10 is a block diagram of exemplary computing device(s) 1000 suitable for use in implementing some embodiments of the present disclosure. Computing device 1000 may include an interconnect system 1002 that directly or indirectly couples the following devices: memory 1004, one or more central processing units (CPUs) 1006, one or more graphics processing units (GPUs) 1008, a communications interface 1010, input / output (I / O) ports 1012, input / output components 1014, a power supply 1016, one or more presentation components 1018 (e.g., display(s)), and one or more logic units 1020. In at least one embodiment, computing device(s) 1000 may include one or more virtual machines (VMs), and / or any of the components thereof may include virtual components (e.g., virtual hardware components).For non-limiting examples, one or more of the GPUs 1008 may include one or more vGPUs, one or more of the CPUs 1006 may include one or more vCPUs, and / or one or more of the logic units 1020 may include one or more virtual logic units. Accordingly, a computing device(s) 1000 may include discrete components (e.g., an entire GPU dedicated to the computing device 1000), virtual components (e.g., a portion of a GPU dedicated to the computing device 1000), or a combination thereof.

[0162] Although the different blocks of Fig. 10 as being connected with lines via the interconnect system 1002, this is not intended to be limiting and is for clarity only. For example, in some embodiments, a presentation component 1018, such as a display device, may be considered an I / O component 1014 (e.g., if the display is a touchscreen). As another example, the CPUs 1006 and / or GPUs 1008 may include memory (e.g., memory 1004 may be representative of a memory device in addition to the memory of the GPUs 1008, the CPUs 1006, and / or other components). In other words, the computing device of Fig. 10 is merely illustrative. No distinction is made between categories such as “workstation,” “server,” “laptop,” “desktop,” “tablet,” “client device,” “mobile device,” “handheld device,” “game console,” “electronic control unit (ECU),” “virtual reality system,” and / or other device or system types, since all fall within the scope of the computing device. Fig. 10 should be considered.

[0163] The interconnect system 1002 may represent one or more connections or buses, such as an address bus, a data bus, a control bus, or a combination thereof. The interconnect system 1002 may include one or more bus or connection types, such as an Industry Standard Architecture (ISA) bus, an Extended Industry Standard Architecture (EISA) bus, a Video Electronics Standards Association (VESA) bus, a Peripheral Component Interconnect (PCI) bus, a Peripheral Component Interconnect Express (PCIe) bus, and / or another type of bus or connection. In some embodiments, there are direct connections between components. As one example, the CPU 1006 may be directly connected to the memory 1004. Further, the CPU 1006 may be directly connected to the GPU 1008. If there is a direct or point-to-point connection between components, the interconnect system 1002 may include a PCIe connection to implement the connection.In these examples, a PCI bus need not be included in the computing device 1000.

[0164] Memory 1004 may include any of a variety of computer-readable media. The computer-readable media may be any available media accessible by the computing device 1000. The computer-readable media may include both volatile and non-volatile media, and removable and non-removable media. By way of example and not limitation, the computer-readable media may include computer storage media and communication media.

[0165] The computer storage media may include both volatile and non-volatile 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 1004 may store computer-readable instructions (e.g., representing a program(s) and / or a program element(s)), such as an operating system.Computer storage media may include, but is not limited to, RAM, ROM, EEPROM, flash memory or other storage technology, CD-ROM, Digital Versatile Disks (DVD) or other optical disk storage, magnetic 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 that can be accessed by the computing device 1000. As used herein, computer storage media does not include signals per se.

[0166] The computer storage media may embody computer-readable instructions, data structures, program modules, and / or other data types in a modulated data signal, such as a carrier wave or other transport mechanism, and include any information transmission media. The term "modulated data signal" may refer to a signal having one or more of its properties adjusted or altered in such a way as to encode information in the signal. By way of example and not limitation, the computer storage media may include wired media, such as a wired network or a direct-wired connection, and wireless media, such as acoustic, RF, infrared, and other wireless media. Combinations of any of the foregoing should also be included within the scope of computer-readable media.

[0167] The CPU(s) 1006 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 1000 to perform one or more of the methods and / or processes described herein. The CPU(s) 1006 may each include one or more cores (e.g., one, two, four, eight, twenty-eight, seventy-two, etc.) capable of handling a plurality of software threads concurrently. The CPU(s) 1006 may include any type of processor and may include different types of processors depending on the type of computing device 1000 implemented (e.g., processors with fewer cores for mobile devices and processors with more cores for servers).For example, depending on the type of computing device 1000, the processor may be an advanced RISC machine (ARM) processor implemented using reduced instruction set computing (RISC) or an x86 processor implemented using complex instruction set computing (CISC). Computing device 1000 may include one or more CPUs 1006 in addition to one or more microprocessors or supplemental coprocessors, such as math coprocessors.

[0168] In addition or alternatively to the CPU(s) 1006, the GPU(s) 1008 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 1000 to perform one or more of the methods and / or processes described herein. One or more of the GPU(s) 1008 may be an integrated GPU (e.g., with one or more of the CPU(s) 1006) and / or one or more of the GPU(s) 1008 may be a discrete GPU. In embodiments, one or more of the GPU(s) 1008 may be a coprocessor of one or more of the CPU(s) 1006. The GPU(s) 1008 may be used by the computing device 1000 to render graphics (e.g., 3D graphics) or to perform general purpose computations. For example, the GPU(s) 1008 can be used for general purpose computing on GPUs (GPGPU).The GPU(s) 1008 may include hundreds or thousands of cores capable of handling hundreds or thousands of software threads simultaneously. The GPU(s) 1008 may generate pixel data for output images in response to rendering commands (e.g., rendering commands from the CPU(s) 1006 received via a host interface). The GPU(s) 1008 may include graphics memory, such as display memory, for storing pixel data or any other suitable data, such as GPGPU data. The display memory may be included as part of the memory 1004. The GPU(s) 1008 may include two or more GPUs operating in parallel (e.g., via an interconnect). The interconnect may connect the GPUs directly (e.g., using NVLINK) or may connect the GPUs through a switch (e.g., using NVSwitch).When combined, each GPU can generate 1008 pixel data or GPGPU data for different sections of an output or for different outputs (e.g., a first GPU for a first image and a second GPU for a second image). Each GPU can contain its own memory or can share memory with other GPUs.

[0169] In addition to or alternatively to the CPU(s) 1006 and / or the GPU(s) 1008, the logic unit(s) 1020 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 1000 to perform one or more of the methods and / or processes described herein. In embodiments, the CPU(s) 1006, the GPU(s) 1008, and / or the logic unit(s) 1020 may discretely or jointly perform any combination of the methods, processes, and / or portions thereof. One or more of the logic units 1020 may be part of and / or integrated with one or more of the CPU(s) 1006 and / or the GPU(s) 1008, and / or one or more of the logic units 1020 may be discrete components or otherwise external to the CPU(s) 1006 and / or the GPU(s) 1008.In embodiments, one or more of the logic units 1020 may be a coprocessor of one or more of the CPU(s) 1006 and / or one or more of the GPU(s) 1008.

[0170] Examples of the logic unit(s) 1020 include one or more processing cores and / or components thereof, such as data processing units (DPUs), tensor cores (TCs), tensor processing units (TPUs), visual pixel cores (PVCs), vision processing units (VPUs), graphics processing clusters (GPCs), texture processing clusters (TPCs), streaming multiprocessors (SMs), tree traversal units (TTUs), artificial intelligence accelerators (AIAs), deep learning accelerators (DLAs), arithmetic logic units (ALUs), application-specific integrated circuits (ASICs), floating-point units (FPUs), input / output (I / O) elements, peripheral component interconnect (PCI) or peripheral component interconnect express (PCIe) elements, and / or the like.

[0171] The communication interface 1010 may include one or more receivers, transmitters, and / or transceivers that enable the computing device 1000 to communicate with other computing devices over an electronic communication network, including wired and / or wireless communication. The communication interface 1010 may include components and functionality for enabling 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 logic unit(s) 1020 and / or the communication interface 1010 may include one or more data processing units (DPUs) for transferring data received over a network and / or through the interconnect system 1002 directly to (e.g., a memory of) one or more GPU(s) 1008.

[0172] The I / O ports 1012 may enable the computing device 1000 to be logically coupled to other devices, including the I / O components 1014, the presentation component(s) 1018, and / or other components, some of which may be built into (e.g., integrated with) the computing device 1000. Illustrative I / O components 1014 include a microphone, a mouse, a keyboard, a joystick, a game pad, a game controller, a satellite dish, a scanner, a printer, a wireless device, etc. The I / O components 1014 may provide a natural user interface (NUI) that processes air gestures, speech, or other physiological inputs generated by a user. In some cases, inputs may be transmitted to a suitable network element for further processing.An NUI may implement any combination of speech recognition, stylus recognition, facial recognition, biometric recognition, both on-screen and off-screen gesture recognition, air gestures, head and eye tracking, and touch detection (as described in more detail below) associated with a display of the computing device 1000. The computing device 1000 may include depth cameras, such as stereoscopic camera systems, infrared camera systems, RGB camera systems, touchscreen technology, and combinations thereof for gesture detection and recognition. Additionally, the computing device 1000 may include accelerometers or gyroscopes (e.g., as part of an inertial measurement unit (IMU)) that enable detection of motion. In some examples, the output of the accelerometers or gyroscopes may be used by the computing device 1000 to render immersive augmented reality or virtual reality.

[0173] Power supply 1016 may include a hardwired power supply, a battery power supply, or a combination thereof. Power supply 1016 may provide power to computing device 1000 to enable the components of computing device 1000 to operate.

[0174] The presentation component(s) 1018 may include a display (e.g., a monitor, a touchscreen, a television monitor, a front-panel display (HUD), other display types, or a combination thereof), speakers, and / or other presentation components. The presentation component(s) 1018 may receive data from other components (e.g., the GPU(s) 1008, the CPU(s) 1006, DPUs, etc.) and output the data (e.g., as an image, video, audio, etc.). EXEMPLARY DATA CENTER

[0175] Fig. Figure 11 illustrates an example data center 1100 that may be used in at least one embodiment of the present disclosure. Data center 1100 may include a data center infrastructure layer 1110, a framework layer 1120, a software layer 1130, and / or an application layer 1140.

[0176] As in Fig. 11, the data center infrastructure layer 1110 may include a resource orchestrator 1112, clustered compute resources 1114, and node compute resources (“node CRs”) 1116(1)-1116(N), where “N” represents any positive whole number. In at least one embodiment, the node CRs 1116(1)-1116(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 memory), storage devices (e.g., solid-state or disk drives), network input / output (NW I / O) devices, network switches, virtual machines (VMs), power modules, and / or cooling modules, etc. In some embodiments, one or more node CRs from the node CRs 1116(1)-1116(N) may correspond to a server having one or more of the above-mentioned computing resources. Additionally, in some embodiments, the node CRs 1116(1)-1116(N) may include one or more virtual components, such as vGPUs, vCPUs, and / or the like, and / or one or more of the node CRs 1116(1)-1116(N) may correspond to a virtual machine (VM).

[0177] In at least one embodiment, the grouped computing resources 1114 may include separate groupings of node CRs 1116 housed in one or more racks (not shown) or multiple racks housed in data centers in different geographic locations (also not shown). Separate groupings of node CRs 1116 within the grouped computing resources 1114 may include grouped computing, networking, storage, or memory resources that may be configured or assigned to support one or more workloads. In at least one embodiment, multiple node CRs 1116 including CPUs, GPUs, DPUs, and / or other processors may be grouped in one or more racks to provide computing resources to support one or more workloads.The one or more racks may also contain any number of power modules, cooling modules, and / or network switches in any combination.

[0178] Resource orchestrator 1112 may configure or otherwise control one or more node CRs 1116(1)-1116(N) and / or clustered compute resources 1114. In at least one embodiment, resource orchestrator 1112 may include a software design infrastructure (SDI) management entity for data center 1100. Resource orchestrator 1112 may include hardware, software, or a combination thereof.

[0179] In at least one embodiment, as in Fig. 11, the framework layer 1120 may include a job scheduler 1133, a configuration manager 1134, a resource manager 1136, and / or a distributed file system 1138. The framework layer 1120 may include a framework to support software 1132 of the software layer 1130 and / or one or more applications 1142 of the application layer 1140. The software 1132 or the application(s) 1142 may each include web-based service software or applications, such as those provided by Amazon Web Services, Google Cloud, and Microsoft Azure. The framework layer 1120 may be some type of free and open-source software web application framework, such as, but not limited to, Apache Spark™ (hereinafter "Spark"), which may utilize the distributed file system 1138 for large-scale data processing (e.g., "Big Data").In at least one embodiment, job scheduler 1133 may include a Spark driver to enable scheduling of workloads supported by various layers of data center 1100. Configuration manager 1134 may be capable of configuring various layers, such as software layer 1130 and framework layer 1120, including Spark and distributed file system 1138, to support large-scale computing. Resource manager 1136 may be capable of managing clustered or grouped compute resources mapped to or allocated to support distributed file system 1138 and job scheduler 1133. In at least one embodiment, clustered or grouped compute resources may include clustered compute resource 1114 in data center infrastructure layer 1110.The resource manager 1136 may coordinate with the resource orchestrator 1112 to manage these mapped or allocated computing resources.

[0180] In at least one embodiment, the software 1132 included in the software layer 1130 may include software used by at least portions of the node CRs 1116(1)-1116(N), the clustered computing resources 1114, and / or the distributed file system 1138 of the framework layer 1120. One or more types of software may include, but are not limited to, Internet website search software, email virus scanning software, database software, and streaming video content software.

[0181] In at least one embodiment, the application(s) 1142 included in the application layer 1140 may include one or more types of applications used by at least portions of the node CRs 1116(1)-1116(N), the clustered compute resources 1114, and / or the distributed file system 1138 of the framework layer 1120. One or more types of applications may include, but are not limited to, any number of a genomic application, a cognitive computing application, and a machine learning application, including training or inference software, machine learning framework software (e.g., PyTorch, TensorFlow, Caffe, etc.), and / or other machine learning applications used in connection with one or more embodiments.

[0182] In at least one embodiment, any of the configuration manager 1134, the resource manager 1136, and the resource orchestrator 1112 may implement any number and type of self-modifying actions based on any amount and type of data collected in any technically feasible manner. Self-modifying actions may relieve a data center operator of the data center 1100 from making potentially poor configuration decisions and potentially avoid underutilized and / or poorly performing parts of a data center.

[0183] Data center 1100 may include tools, services, software, or other resources to train one or more machine learning models or to predict or infer information using one or more machine learning models according to one or more embodiments described herein. For example, one or more machine learning models may be trained by calculating weight parameters according to a neural network architecture using software and / or computational resources described above with respect to data center 1100.In at least one embodiment, trained or deployed machine learning models corresponding to one or more neural networks may be used to infer or predict information using resources described above with respect to data center 1100 by using weight parameters calculated by one or more training techniques, such as, but not limited to, those described herein.

[0184] In at least one embodiment, data center 1100 may utilize CPUs, application-specific integrated circuits (ASICs), GPUs, FPGAs, and / or other hardware (or virtual computing resources corresponding thereto) to perform training and / or inference using resources described above. Furthermore, one or more of the software and / or hardware resources described above may be configured as a service to enable users to train or perform information inference, such as image recognition, speech recognition, or other artificial intelligence services. EXAMPLE NETWORK ENVIRONMENTS

[0185] Network environments suitable for use in implementing embodiments of the 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 based on one or more instances of the computing device(s) 1000 of Fig. 10 - e.g., each device may include similar components, features, and / or functionality of the computing device(s) 1000. Additionally, if backend devices (e.g., servers, NAS, etc.) are implemented, the backend devices may be included as part of a data center 1100, an example of which is described herein with respect to Fig. 11 is described in more detail.

[0186] Components of a network environment can communicate with each other over a network(s), which can be wired, wireless, or both. The network can include multiple networks or a network of 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 a public switched telephone network (PSTN), and / or one or more private networks. If the network includes a wireless telecommunications network, components such as a base station, a communications tower, or even access points (as well as other components) can provide wireless connectivity.

[0187] Compatible network environments may include one or more peer-to-peer network environments—in which case, a server may not be included in a network environment—and one or more client-server network environments—in which case, one or more servers may be included in a network environment. In peer-to-peer network environments, the functionality described herein with respect to one or more servers may be implemented on any number of client devices.

[0188] In at least one embodiment, a network environment may include one or more cloud-based network environments, a distributed computing environment, a combination thereof, etc. 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. A framework layer may include a framework to support software of a software layer and / or one or more applications of an application layer. The software or application(s) may each include web-based service software or applications. In embodiments, one or more of the client devices may use the web-based service software or applications (e.g.,by accessing the service software and / or applications through one or more application programming interfaces (APIs). The framework layer may be some type of free and open-source software web application framework, such as, but not limited to, a distributed file system for large-scale data processing (e.g., "big data").

[0189] A cloud-based network environment may provide cloud computing and / or cloud storage performing any combination of computing and / or data storage functions (or one or more portions thereof) described herein. Each of these various functions may be distributed across multiple locations of central or core servers (e.g., one or more data centers that may be distributed across a state, region, country, globe, etc.). When a connection to a user (e.g., a client device) is relatively close to one or more edge servers, one or more core servers may designate at least a portion of the functionality for the edge server(s). A cloud-based network environment may 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).

[0190] The client device(s) may include at least some of the components, features, and functionality of the devices described herein with respect to Fig.10 described exemplary computing device(s) 1000.By way of example, and not limitation, a 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 headset, a global positioning system (GPS) or device, a video player, a video camera, a surveillance device or system, a vehicle, a boat, a flying vessel, a virtual machine, a drone, a robot, a handheld communication device, a hospital device, a gaming device or system, an entertainment system, a vehicle computing system, an embedded system controller, a remote controller, an appliance, a consumer electronics device, a workstation, an edge device, any combination of these delineated devices, or any other suitable device.

[0191] The disclosure may be described in the general context of computer code or machine-usable instructions, including computer-executable instructions, such as program modules, executed by a computer or other machine, such as a personal data assistant or other handheld device. In general, program modules, including routines, programs, objects, components, data structures, etc., refer to code that performs specific tasks or implements specific abstract data types. The disclosure may be implemented in a variety of system configurations, including handheld devices, consumer electronics, general-purpose computers, specialized computing devices, etc. The disclosure may also be implemented in distributed computing environments where tasks are performed by remote processing devices connected by a communications network.

[0192] As used herein, a recitation of "and / or" with reference to two or more elements should be interpreted to mean 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. Additionally, "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.

[0193] The subject matter of the present disclosure is specifically described herein to satisfy legal requirements. However, the description itself is not intended to limit the scope of this disclosure. Rather, the inventors have contemplated that the claimed subject matter could be embodied in other ways to include various steps or combinations of steps similar to those described herein in connection with other present or future technologies. Furthermore, although the terms "step" and / or "block" may be used herein to refer to various elements of methods employed, the terms should not be interpreted to imply any particular ordering between or among different steps disclosed herein, unless and unless the order of individual steps is explicitly described. Example paragraphs

[0194] A method comprising: generating, based at least on LiDAR data obtained using one or more LiDAR sensors, image data representative of an image corresponding to at least a portion of an environment; determining, using one or more machine learning models and based at least on the image data, a position associated with a road marker within the environment; and causing a map to display the position associated with the road marker within the environment.

[0195] The method of paragraph A, wherein determining the position associated with the road marking within the environment comprises: determining, using the one or more machine learning models and based at least on the image data, a boundary shape indicating a portion of the image depicting the road marking; and determining, based at least on the portion of the image, the position associated with the road marking within the environment.

[0196] The method of paragraph B, wherein causing the map to display the position associated with the road marker within the environment comprises: determining one or more first positions associated with one or more first points of the bounding shape within the image; determining one or more second positions of one or more second points within the map that correspond to the one or more first positions of the one or more first points within the image; and updating, based at least on the one or more second positions of the one or more second points, a portion of the map to display the position associated with the road marker within the environment.

[0197] The method of paragraph B, wherein the boundary shape includes an orientation based on at least a direction of travel of a road for which the road marking is positioned.

[0198] The method of any one of paragraphs AD, further comprising: determining, using the one or more machine learning models and based at least on the image data, a classification associated with the road marking; and causing the map to display the classification associated with the road marking.

[0199] A method according to any one of paragraphs AE, wherein the image comprises at least one of: a top-down image corresponding to the at least one portion of the environment; or a top-down image displaying one or more intensities associated with one or more points corresponding to the at least one portion of the environment.

[0200] The method of any one of paragraphs AF, further comprising: determining a position associated with a machine that generated the LiDAR data using the one or more LiDAR sensors, wherein determining the position associated with the road marking within the environment is further based at least on the position associated with the machine.

[0201] The method of any one of paragraphs AG, further comprising: generating, based at least on the LiDAR data and motion data representing a movement of a machine during the generation of the LiDAR data, point cloud data representing points, wherein the generation of the image data is based at least on the point cloud data.

[0202] A system comprising: one or more processing units for: generating, based at least on sensor data obtained using one or more sensors, image data representative of a top-down image corresponding to at least a portion of an environment; determining, using one or more machine learning models and based at least on the image data, a position associated with a road marking within the environment; and encoding the position associated with the road marking in map data associated with a map corresponding to the portion of the environment.

[0203] The system of paragraph I, wherein determining the position associated with the road marking within the environment comprises: determining, using the one or more machine learning models and based at least on the image data, a boundary shape indicating a portion of the top-down image representing the road marking; and determining, based at least on the portion of the top-down image, the position of the road marking within the environment.

[0204] The system of paragraph J, wherein the position is encoded based at least on: determining that a portion of the map corresponds to a portion of the top-down image associated with the boundary shape; and encoding the position associated with the road marking.

[0205] A system according to paragraph J, wherein the boundary shape includes an orientation based on at least one direction of travel of a road for which the road marking is positioned.

[0206] The system of any one of paragraphs IL, wherein the one or more processing units are further operable to: determine, using the one or more machine learning models and based at least on the image data, a classification associated with the road marking; and encode the classification associated with the road marking into the map data.

[0207] A system according to any one of paragraphs III, wherein the top-down image displays one or more intensities associated with one or more points within at least the portion of the environment.

[0208] The system of any one of paragraphs IN, wherein: the sensor data obtained using the one or more sensors comprises LiDAR data obtained using one or more LiDAR sensors; and the generation of the image data representative of the top-down image corresponding to at least the portion of the environment is based on at least a point cloud associated with the LiDAR data.

[0209] The system of any one of paragraphs 10, wherein the one or more processing units are further operable to: determine a position associated with a machine that generated the sensor data using the one or more sensors, wherein the determination of the position associated with the road marking within the environment is further based at least on the position associated with the machine.

[0210] A system according to any one of paragraphs IP, wherein the system is included in at least one of the following: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system implemented using an edge device; a system implemented using a robot; a system implementing one or more large language models; a system for performing conversational AI operations; a system for generating synthetic data; a system including one or more virtual machines (VMs);a system implemented at least in part in a data center; or a system implemented at least in part using cloud computing resources.

[0211] A processor comprising: one or more processing units to cause one or more operations associated with a machine to be performed based on at least a position associated with a road marking encoded in map data corresponding to a map, wherein the position associated with the road marking is determined using one or more machine learning models and based on at least an image displaying one or more intensity values ​​associated with one or more points represented by LiDAR data, the image representing at least a portion of the environment containing the road marking.

[0212] The processor of paragraph R, wherein the position associated with the road marking is further determined at least in part by: determining, using the one or more machine learning models and based at least on the image, a boundary shape indicating a portion of the image depicting the road marking; and determining, based at least on the portion of the image, the position associated with the road marking within the environment.

[0213] The system of paragraph R or paragraph S, wherein the system is included in at least one of the following: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system implemented using an edge device; a system implemented using a robot; a system implementing one or more large language models; a system for performing conversational AI operations; a system for generating synthetic data; a system including one or more virtual machines (VMs);a system implemented at least in part in a data center; or a system implemented at least in part using cloud computing resources. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] US 16 / 101 232

[0096] Cited non-patent literature

[0000] Society of Automotive Engineers (SAE) “Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles” (Standard No. J3016-201806, published on June 15, 2018, Standard No. J3016-201609, published on September 30, 2016

[0054]

Claims

[1] Method comprising: Generating, based at least on LiDAR data obtained using one or more LiDAR sensors, image data representative of an image corresponding to at least a portion of an environment; Determining, using one or more machine learning models and based at least on the image data, a position associated with a road marking within the environment; and Cause a map to show the position associated with the road marker within the surrounding area. [2] The method of claim 1, wherein determining the position associated with the road marking within the environment comprises: Determining, using the one or more machine learning models and based at least on the image data, a boundary shape indicating a portion of the image representing the road marking; and Determine, based at least on the portion of the image, the position associated with the road marking within the environment. [3] The method of claim 2, wherein causing the map to display the position associated with the road marker within the environment comprises: Determining one or more first positions associated with one or more first points of the boundary shape within the image; Determining one or more second positions of one or more second points within the map that correspond to the one or more first positions of the one or more first points within the image; and Updating, based at least on the one or more second positions of the one or more second points, a portion of the map to indicate the position associated with the road marker within the environment. [4] The method of claim 2, wherein the boundary shape includes an orientation based at least on a direction of travel of a road for which the road marker is positioned. [5] The method of claim 1, further comprising: Determining, using the one or more machine learning models and based at least on the image data, a classification associated with the road marking; and Make the map display the classification associated with the road marker. [6] The method of claim 1, wherein the image comprises at least one of: a top-down image corresponding to at least one section of the environment; or a top-down image displaying one or more intensities associated with one or more points corresponding to the at least one portion of the environment. [7] The method of claim 1, further comprising: Determining a position associated with a machine that generated the LiDAR data using the one or more LiDAR sensors, wherein determining the position associated with the road marking within the environment is further based at least on the position associated with the machine. [8] The method of claim 1, further comprising: Generating, based at least on the LiDAR data and motion data representing a movement of a machine when generating the LiDAR data, point cloud data representing points, whereby the generation of the image data is based at least on the point cloud data. [9] System comprising: one or more processing units for: Generating, based at least on sensor data obtained using one or more sensors, image data representative of a top-down image corresponding to at least a portion of an environment; Determining, using one or more machine learning models and based at least on the image data, a position associated with a road marking within the environment; and Encoding the position associated with the road marker into map data associated with a map corresponding to the section of the surrounding area. [10] The system of claim 9, wherein determining the position associated with the road marking within the environment comprises: Determining, using the one or more machine learning models and based at least on the image data, a boundary shape indicating a portion of the top-down image representing the road marking; and Determine, based on at least the portion of the top-down image, the position of the road marking within the environment. [11] The system of claim 10, wherein the position is encoded based on at least: Determining that a portion of the map corresponds to a portion of the top-down image associated with the boundary shape; and Encoding the position assigned to the road marking. [12] The system of claim 10, wherein the boundary shape includes an orientation based at least on a direction of travel of a road for which the road marker is positioned. [13] The system of claim 9, wherein the one or more processing units are further configured to: Determining, using the one or more machine learning models and based at least on the image data, a classification associated with the road marking; and Encoding the classification associated with the road marking into the map data. [14] The system of claim 9, wherein the top-down image displays one or more intensities associated with one or more points within at least the portion of the environment. [15] The system of claim 9, wherein: the sensor data obtained using the one or more sensors comprises LiDAR data obtained using one or more LiDAR sensors; and the generation of image data representative of the top-down image corresponding to at least the section of the environment is based on at least one point cloud associated with the LiDAR data. [16] The system of claim 9, wherein the one or more processing units are further configured to: Determining a position associated with a machine that generated the sensor data using the one or more sensors, wherein the determination of the position associated with the road marking within the environment is further based at least on the position associated with the machine. [17] The system of claim 9, wherein the system is comprised of at least one of the following: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system implemented using an edge device; a system implemented using a robot; a system that implements one or more large language models; a system for performing conversational AI operations; a system for generating synthetic data; a system that contains one or more virtual machines (VMs); a system that is at least partially implemented in a data center; or a system implemented at least in part using cloud computing resources. [18] Processor comprising: one or more processing units to cause one or more operations associated with a machine to be performed based on at least a position associated with a road marking encoded in map data corresponding to a map, wherein the position associated with the road marking is determined using one or more machine learning models and based on at least an image displaying one or more intensity values ​​associated with one or more points represented by LiDAR data, the image representing at least a portion of the environment containing the road marking. [19] The processor of claim 18, wherein the position associated with the road marking is further determined at least in part by: Determining, using the one or more machine learning models and based at least on the image, a boundary shape indicating a portion of the image representing the road marking; and Determine, based at least on the portion of the image, the position associated with the road marking within the environment. [20] The processor of claim 18, wherein the processor is included in at least one of the following: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system implemented using an edge device; a system implemented using a robot; a system that implements one or more large language models; a system for performing conversational AI operations; a system for generating synthetic data; a system that contains one or more virtual machines (VMs); a system that is at least partially implemented in a data center; or a system implemented at least in part using cloud computing resources.

Images