Particle-based hazard detection for autonomous machine applications

Abstract

The present disclosure relates to particle-based hazard detection for autonomous machine applications. In different examples, a hazard detection system fuses output from multiple sensors over time to determine a probability that a stationary object or hazard exists at a location. The system can then use the sensor data to compute a detection boundary shape for the detected object, and using the boundary shape, a set of particles can be generated, each particle including a confidence value that an object exists at a corresponding location. The system can then capture additional sensor data through one or more sensors of the autonomous machine that are different from the sensors used to capture the first sensor data. To improve the accuracy of the particle confidence, the system can determine a correspondence between the first sensor data and the additional sensor data (e.g., depth sensor data) that can be used to filter out a portion of the particles and improve the depth prediction corresponding to the object.

Description

Background Art

[0001] The ability to safely detect and avoid hazards on the road is a critical task for any autonomous or semi-autonomous driving system. These driving systems typically need to determine the three-dimensional (3D) location of hazards relative to the self-machine, as well as the hazard classification. While some traditional systems are configured for hazard detection, these systems often provide insufficient confidence levels for detection.

[0002] For example, advanced monocular camera detection systems may be able to estimate the size, type, and location of hazards on a two-dimensional (2D) image plane. However, these monocular camera detection systems cannot provide reliably accurate depth estimates. Therefore, these systems are often used only for classification purposes. As another example, while LiDAR systems can provide depth estimates of hazards, these systems often provide sparse samples of small objects that are insufficient to determine the size and shape of the hazard. Furthermore, because the output of these LiDAR systems represents a cluster of 3D points (e.g., a 3D point cloud), these systems are often unable to provide useful hazard classification. Summary of the Invention

[0003] Embodiments of the present disclosure relate to particle-based hazard detection for autonomous machine applications. The disclosed systems and methods can determine the probability of a stationary object or hazard being present at a location.

[0004] Compared to conventional systems, such as the systems described above, the systems and methods of the present disclosure fuse time-varying outputs from multiple sensors to determine the probability that a stationary object or hazard exists at a certain location. For example, as the ego machine travels along a road, the ego machine's system can analyze the captured sensor data to detect static objects. The system can then use the sensor data to calculate the boundary shape of each detected object, and using the boundary shape, a set of particles can be generated, each particle including a confidence value that the object exists at the corresponding location. The system can then capture additional sensor data through one or more sensors of the ego machine that are different from the sensors used to capture the first sensor data. In order to improve the accuracy of the confidence of the particles at time t+1, the system can determine a correspondence between the first sensor data and the additional sensor data (e.g., depth sensor data), which can be used to improve the depth prediction corresponding to the object.

[0005] The correspondences determined between the first sensor data and the additional sensor data can be used to adjust the weights associated with each particle in the set of particles for each detected potential hazard, increasing the weights where a correspondence exists and decreasing the weights where there is little correspondence. Once the weights are updated, a threshold confidence value can be applied to filter out particles with a confidence value below the threshold. The system can repeat these operations over time to continuously resample the driving environment and incorporate additional sensor data. Based on each additional time step and data sample, the system can further filter the particles by adjusting their corresponding weights to produce a higher level of confidence that a hazard is present or absent at the particle's location. Once the system reaches the threshold confidence level, it can output an indication of the detected hazard, which can be used by other systems in the ego machine. Thus, multiple sensor output types can be used to generate a more accurate representation of the location of hazards or static objects in the environment, accounting for the shortcomings of individual sensor types while leveraging the combined strengths of each sensor type. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The present system and method for particle-based hazard detection for autonomous machine applications is described in detail below with reference to the accompanying drawings, in which:

[0007] Figure 1 is an example data flow diagram of a particle-based hazard detection system according to some embodiments of the present disclosure.

[0008] Figure 2 is an example visualization of an object detection process according to some embodiments of the present disclosure.

[0009] Figure 3 is an example visualization of a particle filtering process according to some embodiments of the present disclosure.

[0010] Figures 4 and 5 is a flow chart illustrating a particle-based hazard detection method according to some embodiments of the present disclosure.

[0011] Figure 6A is an illustration of an exemplary autonomous vehicle according to some embodiments of the present disclosure;

[0012] Figure 6B According to some embodiments of the present disclosure Figure 6A Example camera positions and fields of view for an exemplary autonomous vehicle;

[0013] Figure 6C According to some embodiments of the present disclosure Figure 6A A block diagram of an exemplary system architecture of an exemplary autonomous vehicle;

[0014] Figure 6D One or more cloud-based servers according to some embodiments of the present disclosure and Figure 6A a system diagram of exemplary autonomous vehicles communicating with each other;

[0015] Figure 7 is a block diagram of an example computing device suitable for implementing some embodiments of the present disclosure; and

[0016] Figure 8 is a block diagram of an example data center suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Systems and methods related to particle-based hazard detection for autonomous machine applications are disclosed. Although the present disclosure can be directed to an exemplary autonomous vehicle 600 (which can alternatively be referred to herein as“vehicle 600” or“ego vehicle 600,” examples of which are described with respect to Figures 6A-6D For example, the systems and methods described herein can be used with, but are not limited to, non-autonomous vehicles, semi-autonomous vehicles (e.g., in one or more adaptive driver assistance systems (ADAS)), manned and unmanned robots or robotic platforms, warehouse vehicles, off-road vehicles, vehicles coupled with one or more trailers, airships, boats, shuttles, emergency vehicles, motorcycles, electric or motorized bicycles, airplanes, construction vehicles, underwater vehicles, drones, and / or other vehicle types. Moreover, although the present disclosure can be described with respect to detecting hazards for autonomous vehicle applications, this is not meant to be limiting, and the systems and methods described herein can be used for augmented reality, virtual reality, mixed reality, robotics, security and surveillance, autonomous or semi-autonomous machine applications, and / or any other technical space that can use a perception system.

[0018] To address these deficiencies of conventional systems, the systems and methods of the present disclosure fuse outputs from multiple sensors over time to determine a probability that a stationary object or hazard exists at a location. In some embodiments, as the ego machine travels along a roadway, one or more sensors (e.g., cameras, LiDAR sensors, RADAR sensors, ultrasonic sensors, etc.) of the ego machine can capture sensor data representative of a field of view or a field of sensation of the sensor in the driving environment (e.g., including a path of the ego machine and / or an area surrounding the ego machine). The system of the ego machine can then analyze the sensor data to detect static objects. For example, a camera can capture image data, and the system can use object detection techniques— e.g., neural networks, computer vision, machine learning models, etc.— to detect / identify one or more hazards in the image data.

[0019] In some embodiments, the system can calculate a detection boundary shape for each detected object, and using the boundary shape (e.g., in 2D image space) can generate a projected probability representation (e.g., a probability distribution function (PDF)) in 3D world space (e.g., in a top-down bird's-eye view) that represents the probability of the location of the object or hazard. Using the probabilistic representation, the system can generate a set of particles (e.g., 400, 550, 600, etc.) within the representation (and / or adjacent to the representation) and can assign a confidence level to each particle. Thus, the confidence level can indicate whether the object or hazard is located at the location of the particle. Thus, initially (e.g., at time t=0), the system can output a cluster of particles for each detected potential hazard based on a first set of sensor data (e.g., image data).

[0020] In some embodiments, the system can then capture additional sensor data via one or more sensors of the ego-machine that are different from the sensors used to capture the first sensor data. For example, while a camera is used to capture image data at time t, a LiDAR sensor can be used to capture measurements at time t+1 that can be used to generate a 3D point cloud of the LiDAR sensor's field of view. The system can then process the first sensor data and the additional sensor data to align the timestamps of these sensor data, as sensor data from different sensor types may have been captured at different time intervals. For example, the system can use ego-motion to take a particle cluster generated from image data at time t to predict the location of the particle cluster at time t+1. Based on the inherent error in the ego-motion prediction (and / or intrinsic or extrinsic parameters of the ego-machine's sensors), the confidence corresponding to the particle from time t can be updated at time t+1.

[0021] To improve the accuracy of the confidence level of the particle at time t+1, depth sensor data (e.g., from a LiDAR sensor, RADAR sensor, etc.) can be used to improve the depth prediction associated with the object. For example, using the depth sensor data, timestamps corresponding to the depth sensor data and image data, and ego-motion information, depth detections (e.g., RADAR tracks, LiDAR point clouds, etc.) can be projected into 2D image space for comparison with a bounding shape or other object detection representation represented by the image data. Thus, by aligning the timestamps and using ego-motion between the timestamps and intrinsic and / or extrinsic sensor parameters, the depth detection can be compared to the bounding shape at the relevant timestamps. The depth detections can be projected into image space using a probability distribution—e.g., to account for errors in the ego-motion estimate and sensor-related errors—and points, tracks, or other depth detections that fall within the bounding shape can be used as depth information to update the confidence level associated with the particle at time t+1. Thus, if confidence is initially high over a larger surface area, the confidence level at the actual location of the object or hazard may be higher, for example, because the depth information may be more accurate.

[0022] Thus, the determined correspondence between the image data and the depth data can be used to adjust the weights associated with each particle in the particle cluster for each detected potential hazard, increasing the weights where a correspondence exists and decreasing the weights where there is little to no correspondence. As a high-level example, if image data is used to generate a particle cluster for a potential hazard at time t, the system can use the image data and LiDAR data captured at time t+1 to determine the association between the image data and the LiDAR data (e.g., the system can determine which particles intersect with the 3D points of the LiDAR data). Based on this association, the system can update the weights of each particle in the particle cluster. Once the weights are updated, a threshold confidence value can be used to filter out particles below the threshold confidence. As a result, the remaining particles can more accurately represent the location of static objects or hazards in the environment. In this way, the self-machine can maneuver around or avoid the object or hazard.

[0023] In some embodiments, the system can repeat the above operations over time to continuously resample the driving environment and fuse additional sensor data. Based on each additional time step and data sample, the system can further filter the particles by adjusting their corresponding weights to generate a high level of confidence that a hazard is present or absent at the particle's location. Once the system reaches a threshold confidence level, the system can output an indication of the detected hazard, which can be used by other systems of the ego machine. This indication can include the hazard's 3D location (e.g., relative to the ego machine and / or on a high-definition map), the hazard's classification (e.g., obstacles, parked vehicles, and other static objects / structures on the road), one or more confidence levels, and any additional metadata related to the hazard. While the examples above primarily discuss the use of image data from a single camera and a point cloud from a LiDAR sensor, the present disclosure should not be construed as limited to this combination. In some embodiments, the system can fuse any number of combinations of bounding shapes from a monocular camera, silhouettes from segmented images, disparity maps from a stereo camera, 2D images from an imaging radar, thermal images from a thermal imaging camera, and so on. Furthermore, because each output from the various sensors may include a different strength, the system can fuse sensor data from any number of different sensor types to exploit the different strengths and increase the confidence level of the system's hazard detection.

[0024] Reference Figure 1 , Figure 1 is an exemplary data flow diagram of a particle-based hazard detection system 100 according to some embodiments of the present disclosure. It should be understood that this and other arrangements described herein are presented as examples only. Other arrangements and elements (e.g., machines, interfaces, functions, sequences, functional groupings, etc.) may be used in addition to or in place of those shown, and some elements may be omitted entirely. In addition, many of the elements described herein are functional entities that may be implemented as discrete or distributed components, or in conjunction with other components, in any suitable combination and location. The various functions described herein as being performed by the entities may be performed by hardware, firmware, and / or software. For example, the various functions may be performed by a processor executing instructions stored in a memory. In some embodiments, the system 100 may use a processor similar to Figures 6A-6D Example of autonomous vehicle 600, Figure 7 The exemplary computing device 700 and / or Figure 8 The components, features and / or functions of the exemplary data center 800 are implemented.

[0025] Figure 1The data flow diagram 100 includes ego-motion 102, sensor data 110A / B / C (collectively referred to herein as “sensor data 110”), hazard detectors 112A / B / C (collectively referred to herein as “hazard detectors 112”), intermediate fusers 114A / B, temporal fuser 116, and hazard output 118.

[0026] For non-limiting examples, such as where the sensors generating the sensor data 110 are disposed on or otherwise associated with the vehicle, the sensor data 110 may include data generated by, but not limited to, global navigation satellite system (GNSS) sensors 658 (e.g., global positioning system sensors), radar sensors 660, ultrasonic sensors 662, lidar sensors 664, inertial measurement unit (IMU) sensors 666 (e.g., accelerometers, gyroscopes, magnetic compasses, magnetometers, etc.), microphones 696, stereo cameras 668, wide-angle cameras 670 (e.g., fisheye cameras), infrared cameras 672, surround cameras 674 (e.g., 360-degree cameras), long-range and / or medium-range cameras 698, speed sensors 644 (e.g., for measuring the speed of the vehicle 600), and / or other sensor types.

[0027] In operation, as the vehicle 600 travels along a road, one or more sensors of the vehicle 600 (e.g., a camera, a LiDAR sensor, a RADAR sensor, an ultrasonic sensor, etc.) may capture sensor data 110 representing the sensor's field of view or sense of the driving environment (e.g., including the path of the ego machine and / or the area surrounding the ego machine). The sensor data 110 may then be transmitted to the hazard detector 112.

[0028] The hazard detector 112 may then analyze the sensor data 110A to detect static objects. For example, a camera of the vehicle 600 may capture the sensor data 110A (e.g., image data), and the hazard detector 112 may detect / identify one or more hazards in the sensor data 110 based on object detection techniques using neural networks, computer vision, machine learning models, etc. In some embodiments, the hazard detector 112 may process the sensor data 110A to calculate a detection boundary shape for each detected object—e.g., Figure 2 The boundary shape 202.

[0029] Briefly turn to Figure 3, the hazard detector 112 can use the bounding shape to generate a probability representation 302 projected into the 3D world space (e.g., in a top-down bird's-eye view) that represents the probability of the location of the object or hazard. Using the probability representation 302, the hazard detector 112 can generate particles 304 within (and / or adjacent to) the location of the probability representation 302. The hazard detector 112 can assign a confidence level to each particle 304 that can indicate whether the object or hazard is located at the location of the particle in particles 304. Thus, initially (e.g., at time t=0), the hazard detector 112 can output a particle 304 for each detected potential hazard based on the first sensor data 110A (e.g., image data).

[0030] Back to Figure 1 In some embodiments, the system 100 may then capture sensor data 110B via one or more sensors of the vehicle 600 that are different from the sensors used to capture sensor data 110A. For example, while a camera is used to capture sensor data 110A at time t, a LiDAR sensor may be used to capture sensor data 110B at time t+1. Hazard detector 112B may then use sensor data 110B to generate a 3D point cloud of the LiDAR sensor's sensory field. Mid-level fuser 114A may then process sensor data 110A and sensor data 110B to align timestamps corresponding to sensor data 110, for example, because sensor data 110 from different sensor types may have been captured at different time intervals. For example, the system may use ego-motion 102 to retrieve particle 304 generated by hazard detector 112A at time t to predict the position of particle 304 at time t+1. Based on the inherent error in the ego-motion prediction (and / or intrinsic or extrinsic parameters of the sensors of the vehicle 600 ), the confidence corresponding to the particle 304 from time t may be updated at time t+1.

[0031] In some embodiments, to improve the accuracy of the confidence level of the particle at time t+1, the sensor data 110B may include sensor data representing depth information (e.g., from a LiDAR sensor, a RADAR sensor, etc.). This sensor data 110B may be used to improve the depth prediction corresponding to the object. For example, and briefly turning to Figure 2, using the sensor data 110B, the timestamps corresponding to the sensor data 110B and the sensor data 110A, and the ego-motion 102, depth detections 204A / B / C / D (collectively referred to herein as "depth detections 204") can be projected into the 2D image space 200 for comparison with the boundary shape 202 or other object detection representations represented by the sensor data 110A. Thus, by aligning the timestamps, and using the ego-motion 102 between the timestamps and the intrinsic and / or extrinsic sensor parameters, the depth detection 204 can be compared with the boundary shape 202 at the relevant timestamps. The depth detection 204 can be projected into the image space 200 using a probability distribution, for example, to account for errors in the ego-motion estimate as well as sensor-related errors. As Figure 2 As shown, depth detections 204A and 204B—or at least portions of detections 204A and 204B—fall within bounding shape 202. Consequently, depth information associated with depth detections 204A and 204B may be used to increase the confidence values ​​associated with 204A and 204B. Furthermore, because depth detections 204C and 204D do not fall within bounding shape 202, the confidence values ​​associated with depth detections 204C and 204D may be decreased. In some embodiments, the depth value associated with depth detection 204 may be increased or decreased based solely on whether depth detection 204 falls (at least partially) within bounding shape 202. Furthermore, the depth value may be adjusted based on the distance of each depth detection 204 from bounding shape 202. Thus, the determined correspondence between sensor data 110A (e.g., image data) and sensor data 110B (e.g., depth data) may be processed by the mid-level fuser 114A to adjust the weight associated with each particle in each detected potentially dangerous particle cluster, increasing the weight where the mid-level fuser 114A determines that a correspondence exists and decreasing the weight where the mid-level fuser 114A determines that there is little to no correspondence.

[0032] In some embodiments, the system can repeat the operations described herein over time to continuously resample the driving environment and fuse additional sensor data 110 by using the temporal fusion process of the temporal fuser 116. Based on each additional time step and data sample, the system can further filter the particles by adjusting their corresponding weights to generate a high level of confidence that a hazard is present or absent at the particle's location. Figure 3As a high-level example, image data may be used to determine a probability distribution 302 of a potential hazard at time t. Probability distribution 302 may then be used to generate a particle cluster 304 of the potential hazard at time t, where each particle may correspond to a confidence value (e.g., a weight). Ego-motion data may then be used to predict the location of particle cluster 304 at t+1, represented as particle cluster 308. Image data and LiDAR data captured at t+1 may then be used to determine a probability distribution 306 (similar to the probability distribution for the probability distribution 306) based on the association determined between the image data and the LiDAR data. Figure 2 , described in an embodiment). Using probability distribution 306, the system can update the weight of each particle in particle cluster 308 based on the position of each particle in particle cluster 308 relative to probability distribution 306. For example, the weight associated with particle 310B can be increased relative to particle 310A because particle 310B is within probability distribution 306 compared to particle 310A, which is outside of the probability distribution. Once the weight of each particle in particle cluster 308 is updated, a threshold confidence value can be used to filter out a portion of particle cluster 308 that is below the threshold confidence value to generate filtered particle cluster 312. As a result, filtered particle cluster 312 can more accurately represent the location of static objects or hazards in the environment.

[0033] In some embodiments, additional sensor data types—e.g., sensor data 110C, which may include different sensor data types than sensor data 110A and 110B—may be used by temporal fuser 116 in the temporal fusion process. For example, sensor data 110C (e.g., stereo image data) may be used by hazard detector 112C to generate an indication (e.g., a boundary shape, a particle, a probability distribution, etc.) corresponding to a hazard detected in sensor data 110C. This indication may then be fused by mid-level fuser 114B with outputs from hazard detectors 112B and / or 112A to determine confidence values ​​for particles corresponding to the detected hazard. These particles and confidence values ​​may then be used to further adjust the particle confidence values ​​as part of spatiotemporal fusion process 116.

[0034] Back to Figure 1 Once the system 100 has fused enough sensor data 110 through the processes described herein to reach a threshold confidence level, a hazard output 118 may be generated as an indication of a detected hazard that may be used by other systems of the vehicle 600, such as an obstacle avoidance system, an automatic braking system, a map health system, etc. The hazard output 118 may include the 3D location of the hazard (e.g., relative to the vehicle 600 and / or on a high-definition map), the classification of the hazard (e.g., an obstacle in the road, a parked vehicle, and other static objects / structures), one or more confidence levels, and any additional metadata associated with the hazard.

[0035] Now refer to Figure 4 and Figure 5 , each block of methods 400 and 500 described herein includes a computing process that can be performed using any combination of hardware, firmware, and / or software. For example, the various functions can be performed by a processor executing instructions stored in a memory. Methods 400 and 500 can also be embodied as computer-usable instructions stored on a computer storage medium. Methods 400 and 500 can be provided by a stand-alone application, a service or a hosted service (stand-alone or in combination with another hosted service), or a plug-in to another product, to name a few. Furthermore, methods 400 and 500 are provided by way of example only, with respect to Figure 1 However, the methods may additionally or alternatively be performed by any one system or any combination of systems, including but not limited to those described herein.

[0036] Figure 4 4 is a flow chart illustrating a method 400 for particle-based hazard detection according to some embodiments of the present disclosure. The method 400, at block B402, includes detecting objects in an environment based on at least first sensor data representing an image generated at a first time using an image sensor of the ego machine. For example, while the ego machine is traveling along a road, the ego machine's system may analyze the captured sensor data to detect static objects.

[0037] Method 400, at block B404, includes determining, based at least in part on the detection of the object, a first probability distribution corresponding to the location of the object in a three-dimensional (3D) world space. For example, the system may calculate a detection boundary shape for each detected object and, using the boundary shape (e.g., in a 2D image space), may generate a projected probability representation (e.g., a probability distribution function (PDF)) in a 3D world space (e.g., in a top-down bird's-eye view), the projected probability representation representing the probability of the location of the object or hazard.

[0038] Method 400, at block B406, includes generating a set of particles at a first location within a first probability distribution, the particles having a first confidence value indicating whether an object exists at the first location. For example, using a probabilistic representation, the system can generate a set of particles (e.g., 400, 500, 600, etc.) within (and / or adjacent to) the representation and can assign a confidence value to each particle.

[0039] Method 400, at block 408, includes generating, at a second time after the first time, a particle set for a second location based at least in part on the first location and the motion of the ego-machine between the first time and the second time, the particle set for the second location including a second confidence value indicating whether the object is present at the second location. For example, the system may use the ego-motion to project a particle cluster generated from image data at time t to a future location of the particle cluster at time t+1. Based on inherent errors in the ego-motion prediction (and / or intrinsic or extrinsic parameters of the ego-machine's sensors), the confidence corresponding to the particle from time t may be updated at time t+1.

[0040] Method 400, at block B410, includes correlating the first sensor data with second sensor data representing one or more depth detections generated using a depth sensor of the ego machine at a third time. For example, the system may repeat the above operations over time to continuously resample the driving environment and fuse additional sensor data.

[0041] Method 400, at block B412, includes updating the second confidence value for the set of particles to a third confidence value indicating whether the object is present at the second location based at least in part on the correlation. For example, depth detection from a depth sensor can be used to update the confidence of particles predicted using sensor data from a sensor type that is less accurate in depth prediction (e.g., a camera).

[0042] Method 400, at block B414, includes filtering out one or more of the set of particles having an associated third confidence value less than a threshold confidence value to determine a subset of the particles. For example, based on each additional time step and data sample, the system can further filter the particles by adjusting their corresponding weights to produce a higher level of confidence that a hazard is present or absent at the location of the particles.

[0043] Figure 5 is a flow chart illustrating a method 500 for particle-based hazard detection according to some embodiments of the present disclosure. The method 500, at block B502, includes determining a second particle position corresponding to an object at a second time using a first particle position determined using first sensor data generated using an image sensor and a motion of the ego machine between the first time and the second time, the first particle position being determined using first sensor data generated using an image sensor to predict a particle cluster generated from image data at time t to predict a location of the particle cluster at time t+1. The confidence corresponding to the particle from time t can be updated at time t+1 based on an inherent error in the ego motion prediction (and / or intrinsic or extrinsic parameters of the ego machine's sensors).

[0044] Method 500 , at block B 504 , includes correlating first sensor data generated using the image sensor with second sensor data generated using the depth sensor. For example, the system may repeat the above operations over time to continuously resample the driving environment and fuse additional sensor data.

[0045] Method 500, at block B506, includes comparing, based at least in part on the correlation, one or more depth detections from the second sensor data to a bounding shape corresponding to an object represented by the first sensor data. For example, using the depth sensor data, timestamps corresponding to the depth sensor data and the image data, and ego-motion information, the depth detections (e.g., RADAR tracks, LiDAR point clouds, etc.) can be projected into 2D image space for comparison with the bounding shape or other object detection representation represented by the image data.

[0046] Method 500, at block B508, includes determining a set of one or more depth detections that are within the bounding shape. For example, the depth detections can be projected into image space using a probability distribution—e.g., to account for errors in ego-motion estimation and sensor-related errors—and points, tracks, or other depth detections that fall within the bounding shape can be used as depth information for updating a confidence associated with the particle at time t+1.

[0047] Method 500, at block B510, includes updating the first confidence level to a second confidence level based at least in part on the set of one or more depth detections. For example, the determined correspondence between the image data and the depth data may be used to adjust a weight associated with each particle in the particle cluster for each detected potential hazard, increasing the weight where a correspondence exists and decreasing the weight where little correspondence exists.

[0048] Method 500, at block B512, includes determining the location of the object using a second confidence level. For example, once the system reaches a threshold confidence level, the system can output an indication of the detected hazard, which can be used by other systems of the ego machine. The indication can include the 3D location of the hazard (e.g., relative to the ego machine and / or on a high-definition map), the classification of the hazard (e.g., an obstacle on the road, a parked vehicle, and other static objects / structures), one or more confidence levels, and any additional metadata associated with the hazard.

[0049] Example autonomous vehicle

[0050] Figure 6Ais an illustration of an exemplary autonomous vehicle 600 according to some embodiments of the present invention. Autonomous vehicle 600 (or referred to herein as "vehicle 600") may include, but is not limited to, passenger vehicles such as cars, trucks, buses, first responder vehicles, shuttles, electric or motorized bicycles, motorcycles, fire trucks, police cars, ambulances, boats, construction vehicles, underwater vehicles, drones, vehicles coupled to trailers, and / or other types of vehicles (e.g., vehicles that are unmanned and / or accommodate one or more passengers). Autonomous vehicles are generally described by their level of automation, as defined by the National Highway Traffic Safety Administration (NHTSA) of the U.S. Department of Transportation and the Society of Automotive Engineers (SAE) "Classification and Definitions of Terminology Relating to Automated Driving Systems for Road Motor Vehicles" (Standard No.: J3016-201806, published on June 15, 2018, Standard No.: J3016-201609, published on September 30, 2016, and previous and future versions of this standard). Vehicle 600 may implement functionality according to one or more of the levels 3 through 5 of automated driving. For example, depending on the embodiment, the vehicle 600 may have conditional automation (level 3), high automation (level 4), and / or full automation (level 5).

[0051] Vehicle 600 may include a chassis, a body, wheels (e.g., 2, 4, 6, 8, 18, etc.), tires, axles, and other vehicle components. Vehicle 600 may include a propulsion system 650, such as an internal combustion engine, a hybrid power plant, an all-electric engine, and / or other propulsion system types. Propulsion system 650 may be connected to a drive train of vehicle 600, which may include a transmission, to achieve propulsion of vehicle 600. Propulsion system 650 may be controlled in response to signals received from throttle / accelerator 652.

[0052] When propulsion system 650 is operating (e.g., when the vehicle is in motion), a steering system 654 including a steering wheel can be used to guide vehicle 600 (e.g., along a desired path or route). Steering system 654 can receive signals from steering actuator 656. For fully automated (Level 5) functionality, a steering wheel can be optional.

[0053] Brake sensor system 646 may be used to operate vehicle brakes in response to signals received from brake actuator 648 and / or brake sensors.

[0054] The one or more controllers 636 may include one or more system on chip (SoC) 604 ( Figure 6C) and / or one or more GPUs may provide signals (e.g., representing commands) to one or more components and / or systems of vehicle 600. For example, one or more controllers may send signals to operate vehicle brakes via one or more brake actuators 648, operate steering system 654 via one or more steering actuators 656, and operate propulsion system 650 via one or more throttles / accelerators 652. One or more controllers 636 may include one or more onboard (e.g., integrated) computing devices (e.g., supercomputers) that process sensor signals and output operational commands (e.g., signals representing commands) to enable autonomous driving and / or assist a human driver in driving vehicle 600. One or more controllers 636 may include a first controller 636 for autonomous driving functionality, a second controller 636 for functional safety functionality, a third controller 636 for artificial intelligence functionality (e.g., computer vision), a fourth controller 636 for infotainment functionality, a fifth controller 636 for redundancy in emergency situations, and / or other controllers. In some examples, a single controller 636 may handle two or more of the above functions, two or more controllers 636 may handle a single function, and / or any combination thereof.

[0055] The one or more controllers 636 may provide signals for controlling one or more components and / or systems of the vehicle 600 in response to sensor data (eg, sensor input) received from one or more sensors. Sensor data may be received from, for example, but not limited to, one or more global navigation satellite system sensors 658 (e.g., one or more global positioning system sensors), one or more radar sensors 660, one or more ultrasonic sensors 662, one or more lidar sensors 664, one or more inertial measurement unit (IMU) sensors 666 (e.g., one or more accelerometers, one or more gyroscopes, one or more magnetic compasses, magnetometers, etc.), one or more microphones 696, one or more stereo cameras 668, one or more wide-angle cameras 670 (e.g., fisheye cameras), one or more infrared cameras 672, one or more surround cameras 674 (e.g., 360-degree cameras), one or more long-range and / or mid-range cameras 698, one or more speed sensors 644 (e.g., for measuring the speed of the vehicle 600), one or more vibration sensors 642, one or more steering sensors 640, one or more brake sensors (e.g., as part of a brake sensor system 646), and / or other sensor types.

[0056] One or more of the controllers 636 may receive input (e.g., represented by input data) from the instrument cluster 632 of the vehicle 600 and provide output (e.g., represented by output data, display data, etc.) via a human machine interface (HMI) display 634, audio annunciators, speakers, etc. and / or via other components of the vehicle 600. The output may include information such as vehicle speed, velocity, time, map data (e.g., Figure 6C The HMI display 634 may include information such as the HD map 622 of the vehicle 600, position data (e.g., the position of the vehicle 600, such as the position on the map), direction, the positions of other vehicles (e.g., occupancy grid), and information about objects and object states sensed by the one or more controllers 636. For example, the HMI display 634 may display information about the presence of one or more objects (e.g., street signs, warning signs, traffic light changes, etc.), and / or information about driving maneuvers that the vehicle has performed, is performing, or will perform (e.g., changing lanes now, taking exit 34B in two miles, etc.).

[0057] The vehicle 600 also includes a network interface 624 that can communicate over one or more networks using one or more wireless antennas 626 and / or a modem. For example, the network interface 624 can be capable of communicating over LTE, WCDMA, UMTS, GSM, CDMA2000, etc. The one or more wireless antennas 626 can also enable communication between objects in the environment (e.g., vehicles, mobile devices, etc.) using one or more local area networks (e.g., Bluetooth, Bluetooth LE, Z-wave, ZigBee, etc. and / or low-power wide area networks (LPWANs) such as LoRaWAN, SigFox, etc.).

[0058] Figure 6B According to some embodiments of the present invention Figure 6A 6. Example camera positions and fields of view for an exemplary autonomous vehicle 600. 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 located at different locations on vehicle 600.

[0059] The camera type of the camera may include, but is not limited to, a digital camera, which may be applicable to components and / or systems of the vehicle 600. One or more cameras may operate under Automotive Safety Integrity Level (ASIL) B and / or other ASILs. Depending on the embodiment, the camera type may have any image capture rate, such as 60 frames per second (fps), 120 frames, 240 frames, etc. The camera may use a rolling shutter, a global shutter, another type of shutter, or a combination thereof. In some examples, the color filter array may include a red transparent (RCCC) filter array, a red transparent blue (RCCB) filter array, a red blue green transparent (RBGC) filter array, a Foveon X3 filter array, a Bayer sensor (RGGB) filter array, a monochrome sensor filter array, and / or other types of filter arrays. In some embodiments, a clear pixel camera, such as a camera with an RCCC, RCCB, and / or RBGC filter array, may be used in an effort to improve light sensitivity.

[0060] In some examples, one or more cameras can be used to perform advanced driver assistance system (ADAS) functions (e.g., as part of a redundant or fail-safe design). For example, a multifunctional single camera can be installed to provide functions such as lane departure warning, traffic sign assistance, and intelligent headlight control. One or more cameras (e.g., all cameras) can simultaneously record and provide image data (e.g., video).

[0061] One or more cameras can be mounted in a mounting assembly, such as a custom-designed (3D-printed) assembly, to cut out stray light and interior reflections (e.g., dashboard reflections off the windshield mirror) that could interfere with the camera's ability to capture image data. With respect to the wing mirror mounting assembly, the wing mirror assembly can be custom-3D-printed so that the camera mounting plate matches the shape of the wing mirror. In some examples, one or more cameras can be integrated into the wing mirror. For side-view cameras, one or more cameras can also be integrated into the four pillars at each corner of the cabin.

[0062] A camera (e.g., a front-facing camera) having a field of view that includes a portion of the environment in front of the vehicle 600 can be used for a surround view to help identify the forward path and obstacles, as well as assist in providing information critical to generating an occupancy grid and / or determining a preferred vehicle path with the help of one or more controllers 636 and / or a control SOC. A front-facing camera can be used to perform many of the same ADAS functions as a lidar, including emergency braking, pedestrian detection, and collision avoidance. A forward-facing camera can also be used for ADAS functions and systems, including lane departure warning (LDW), automatic cruise control (ACC), and / or other functions, such as traffic sign recognition.

[0063] Various cameras can be used in the front-facing configuration, including, for example, a monocular camera platform including a CMOS (complementary metal oxide semiconductor) color imager. Another example could be a wide-angle camera 670 that can be used to sense objects entering the field of view from the periphery (e.g., pedestrians, crossing traffic, or bicycles). Although Figure 6B Only one wide-angle camera is shown, but there may be any number of wide-angle cameras 670 on the vehicle 600. In addition, one or more remote cameras 698 (e.g., a long-angle stereo camera pair) can be used for depth-based object detection, especially for objects for which a neural network has not yet been trained. One or more remote cameras 698 can also be used for object detection and classification and basic target tracking.

[0064] One or more stereo cameras 668 may also be included in the front configuration. One or more stereo cameras 668 may include an integrated control unit including an extensible processing unit that may provide programmable logic (FPGA) and a multi-core microprocessor with an integrated CAN or Ethernet interface on a single chip. The unit may be used to generate a three-dimensional map of the vehicle environment, including distance estimates for all points in the image. One or more alternative stereo cameras 668 may include a compact stereo vision sensor that may include two camera lenses (one on the left and one on the right) and an image processing chip that may measure the distance from the vehicle to the target object and use the generated information (e.g., metadata) to activate automatic emergency braking and lane departure warning functions. In addition to the stereo cameras described herein, or alternatively, other types of stereo cameras 668 may be used.

[0065] Cameras with a field of view that includes portions of the surroundings to the sides of the vehicle 600 (e.g., side-view cameras) may be used for surround views, providing information for creating and updating occupancy grids and generating side collision warnings. For example, one or more surround cameras 674 (e.g., Figure 6B Four surround cameras 674 (shown) can be positioned on the vehicle 600. The one or more surround cameras 674 can include one or more wide-angle cameras 670, one or more fisheye cameras, one or more 360-degree cameras, etc. For example, four fisheye cameras can be located on the front, rear, and sides of the vehicle. In an alternative arrangement, the vehicle can use three surround cameras 674 (e.g., left, right, and rear) and can utilize one or more other cameras (e.g., a front-facing camera) as a fourth surround view camera.

[0066] A camera having a field of view that includes a portion of the environment behind the vehicle 600 (e.g., a rearview camera) can be used for parking assistance, surround view, rear-end collision warning, and creating and updating an occupancy grid. A variety of cameras can be used, including but not limited to cameras that are also suitable as front-facing cameras (e.g., one or more long-range and / or mid-range cameras 698, one or more stereo cameras 668, one or more infrared cameras 672, etc.), as described herein.

[0067] Figure 6C According to some embodiments of the present invention Figure 6A 6. Block diagram of an exemplary system architecture for an exemplary autonomous vehicle 600. It will be understood that this and other arrangements described herein are set forth by way of example only. Other arrangements and elements (e.g., machines, interfaces, functions, sequences, functional groupings, etc.) may be used in addition to or in place of the arrangements and elements shown, and some elements may be omitted entirely. In addition, many of the elements described herein are functional entities that may be implemented as discrete or distributed components, or in combination with other components, and in any suitable combination and location. The various functions described herein as being performed by entities may be performed by hardware, firmware, and / or software. For example, the various functions may be performed by a processor executing instructions stored in a memory.

[0068] Figure 6C Each component, feature, and system of the vehicle 600 is connected via a bus 602. The bus 602 may include a controller area network (CAN) data interface (also referred to herein as a "CAN bus"). The CAN may be a network within the vehicle 600 that helps control various features and functions of the vehicle 600, such as the actuation of brakes, acceleration, braking, steering, windshield wipers, and the like. The CAN bus can be configured to have dozens or even hundreds of nodes, each with its own unique identifier (e.g., a CAN ID). The CAN bus can be read to find steering wheel angle, ground speed, engine speed per minute (RPM), button position, and / or other vehicle status indicator lights. The CAN bus may comply with ASIL B standards.

[0069] Although bus 602 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 in addition to the CAN bus. Furthermore, although bus 602 is represented using a single line, this is not intended to be limiting. For example, there may be any number of buses 602, including one or more CAN buses, one or more FlexRay buses, one or more Ethernet buses, and / or one or more other types of buses using different protocols. In some examples, two or more buses 602 may be used to perform different functions and / or for redundancy. For example, a first bus 602 may be used for collision avoidance functionality, and a second bus 602 may be used for drive control. In any example, each bus 602 may communicate with any component of vehicle 600, and two or more buses 602 may communicate with the same component. In some examples, each SoC 604, each controller 636, and / or each computer within the vehicle may have access to the same input data (e.g., input from sensors on vehicle 600) and may be connected to a common bus, such as a CAN bus.

[0070] The vehicle 600 may include one or more controllers 636, as described herein with respect to Figure 6A The controller 636 can be used for various functions. One or more controllers 636 can be coupled to any of the various other components and systems of the vehicle 600 and can be used to control the vehicle 600, the artificial intelligence of the vehicle 600, the infotainment of the vehicle 600, etc.

[0071] The vehicle 600 may include one or more system on a chip (SoC) 604. The SoC 604 may include one or more CPUs 606, one or more GPUs 608, one or more processors 610, one or more caches 612, one or more accelerators 614, one or more data stores 616, and / or other components and features not shown. The one or more SoCs 604 may be used to control the vehicle 600 in various platforms and systems. For example, the one or more SoCs 604 may be combined with an HD map 622 in a system (e.g., a system of the vehicle 600), and the HD map 622 may be downloaded from one or more servers (e.g., a server) via a network interface 624. Figure 6D server 678) to obtain map refreshes and / or updates.

[0072] The one or more CPUs 606 can include a CPU cluster or CPU complex (or referred to herein as a “CCPLEX”). The one or more CPUs 606 can include multiple cores and / or a level two cache. For example, in some embodiments, the one or more CPUs 606 can include eight cores in a coherent multi-processor configuration. In some embodiments, the one or more CPUs 606 can include four dual-core clusters with each cluster having a dedicated level two cache (e.g., a 2 MB level two cache). The one or more CPUs 606 (e.g., CCPLEX) can be configured to support simultaneous cluster operation such that any combination of clusters of CPUs 606 are active at any given time.

[0073] The one or more CPUs 606 can implement power management capabilities including one or more of the following: a single hardware block can be automatically clock-gated when idle to save dynamic power; each core clock can be gated when the core is not actively executing instructions due to execution of WFI / WFE instructions; each core can be independently power-gated; each core cluster can be independently clock-gated when all cores are clock-gated or power-gated; and / or each core cluster can be independently power-gated when all cores are power-gated. The one or more CPUs 606 can further implement enhanced algorithms for managing power states in which allowed power states and expected wake-up times are specified and hardware / microcode determines the best power state for the core, cluster, and CCPLEX to enter. The processing core can support a simplified power state input sequence in software and offload work to microcode.

[0074] The one or more GPUs 608 can include an integrated GPU (or referred to herein as an “iGPU”). The GPU 608 can be programmable and can be efficient on parallel workloads. In some examples, the one or more GPUs 608 can use an enhanced tensor instruction set. The one or more GPUs 608 can include one or more streaming microprocessors, where each streaming microprocessor can include a level one cache (e.g., a level one cache having at least 96 KB of storage capacity) and two or more streaming microprocessors can share a level two cache (e.g., a level two cache having 512 KB of storage capacity). In some embodiments, the one or more GPUs 608 can include at least eight streaming microprocessors. The one or more GPUs 608 can use one or more computing application programming interfaces (APIs). Further, the one or more GPUs 608 can use one or more parallel computing platforms and / or programming models (e.g., NVIDIA’s CUDA).

[0075] One or more GPUs 608 can be power optimized to achieve optimal performance in automotive and embedded use cases. For example, one or more GPUs 608 can be manufactured on fin field-effect transistors (FinFETs). However, this is not intended to be limiting, and one or more GPUs 608 can be manufactured using other semiconductor manufacturing processes. Each streaming microprocessor can incorporate multiple mixed-precision processing cores divided into multiple blocks. For example, but not limited to, 64 PF32 cores and 32 PF64 cores can be divided into four processing blocks. In such an example, each processing block can be allocated 16 FP32 cores, 8 FP64 cores, 16 INT32 cores, two mixed-precision NVIDIA tensor cores for deep learning matrix algorithms, an L0 instruction cache, a warp scheduler, a dispatch unit, and / or a 64KB register file. In addition, the streaming microprocessor may include independent parallel integer and floating-point data paths to provide efficient execution of workloads through mixed and addressed computations. The streaming microprocessor may include independent thread scheduling capabilities to enable finer-grained synchronization and collaboration between parallel threads. Streaming microprocessors may include a combined L1 data cache and shared memory unit to improve performance while simplifying programming.

[0076] One or more GPUs 608 may include high bandwidth memory (HBM) and / or a 16GB HBM2 memory subsystem to provide approximately 900GB / s peak memory bandwidth in some examples. In some examples, synchronous graphics random access memory (SGRAM), such as fifth generation graphics double data rate synchronous random access memory (GDDR5), is used in addition to or alternatively from the HBM memory.

[0077] The fifth generation GPUs 608 may include unified memory technology that includes access counters to allow more accurate migration of memory pages to the processor that most frequently accesses them, thereby improving the efficiency of shared memory ranges between processors. In some examples, address translation services (ATS) support may be used to allow one or more GPUs 608 to directly access the page tables of one or more CPUs 606. In such an example, when the memory management unit (MMU) of one or more GPUs 608 experiences a miss, an address translation request may be sent to the one or more CPUs 606. In response, the one or more CPUs 606 may look up the virtual-to-physical mapping of the address in its page table and send the translation back to the one or more GPUs 608. Thus, unified memory technology may allow a single unified virtual address space for the memory of both the one or more CPUs 606 and the one or more GPUs 608, thereby simplifying programming the one or more GPUs 608 and porting applications to the one or more GPUs 608.

[0078] Additionally, one or more GPUs 608 may include access counters that can track how often one or more GPUs 608 access the memory of other processors. The access counters can help ensure that memory pages are moved to the physical memory of the processor that accesses the page most frequently.

[0079] One or more SoCs 604 may include any number of caches 612, including the caches 612 described herein. For example, one or more caches 612 may include an L3 cache available to one or more CPUs 606 and one or more GPUs 608 (e.g., connecting both one or more CPUs 606 and one or more GPUs 608). Cache 612 may include a write-back cache that can track the state of a line, for example, by using a cache coherence protocol (e.g., MEI, MESI, MSI, etc.). Although smaller cache sizes may be used, according to embodiments, the L3 cache may include 4MB or more.

[0080] One or more SoCs 604 may include one or more arithmetic logic units (ALUs) that can be used to perform processing related to various tasks or operations of the vehicle 600, such as processing a DNN. In addition, one or more SoCs 604 may include one or more floating point units (FPUs) or other math coprocessor or digital coprocessor types for performing mathematical operations within the system. For example, one or more SoCs 104 may include one or more FPUs integrated as execution units within the CPU 606 and / or GPU 608.

[0081] One or more SoCs 604 may include one or more accelerators 614 (e.g., hardware accelerators, software accelerators, or a combination thereof). For example, one or more SoCs 604 may include a hardware acceleration cluster that may include optimized hardware accelerators and / or large on-chip memory. Large on-chip memory (e.g., 4MB of SRAM) enables the hardware acceleration cluster to accelerate neural networks and other calculations. The hardware acceleration cluster can be used to supplement one or more GPUs 608 and offload some tasks of one or more GPUs 608 (e.g., freeing up more cycles of one or more GPUs 608 to perform other tasks). For example, one or more accelerators 614 can be used for target workloads that are stable enough to be suitable for acceleration (e.g., perception, convolutional neural networks (CNNs), etc.). The term "CNN" used herein may include all types of CNNs, including region-based or region-based convolutional neural networks (RCNNs) and fast RCNNs (e.g., for target detection).

[0082] One or more accelerators 614 (e.g., a hardware acceleration cluster) may include one or more deep learning accelerators (DLAs). One or more DLAs may include one or more tensor processing units (TPUs) that may be configured to provide an additional 10 trillion operations per second for deep learning applications and reasoning. A TPU may be an accelerator configured to perform image processing functions (e.g., for CNN, RCNN, etc.) and optimized for them. One or more DLAs may also be optimized for specific neural network types and floating-point operations and reasoning. The design of one or more DLAs may provide higher performance per millimeter than a general-purpose GPU and significantly exceed the performance of a CPU. One or more TPUs may perform multiple functions, including single-instance convolution functions, for example, supporting INT8, INT16, and FP16 data types for features and weights, and post-processor functions.

[0083] One or more DLAs may quickly and efficiently execute neural networks, particularly CNNs, on processed or unprocessed data for any of a variety of functions, including but not limited to: a CNN for object recognition and detection using data from a camera sensor; a CNN for distance estimation using data from a camera sensor; a CNN for emergency vehicle detection and identification and detection using microphone data; a CNN for facial recognition and vehicle owner identification using data from a camera sensor; and / or a CNN for safety and / or security-related events.

[0084] One or more DLAs can perform any function of one or more GPUs 608. For example, by using an inference accelerator, designers can target any function to one or more DLAs or one or more GPUs 608. For example, designers can focus the processing of CNNs and floating-point operations on one or more DLAs and leave other functions to one or more GPUs 608 and / or one or more other accelerators 614.

[0085] One or more accelerators 614 (e.g., a hardware acceleration cluster) may include a programmable vision accelerator (PVA), which may also be referred to herein as a computer vision accelerator. One or more PVAs 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. One or more PVAs may provide a balance between performance and flexibility. For example, each PVA may include, for example, but not limited to, any number of reduced instruction set computer (RISC) cores, direct memory access (DMA), and / or any number of vector processors.

[0086] The RISC core can interact with an image sensor (e.g., an image sensor of any camera described herein), one or more image signal processors, and the like. Each RISC core can include any amount of memory. The RISC core can use any of a variety of protocols, depending on the embodiment. In some examples, the RISC core can execute a real-time operating system (RTOS). The RISC core can be implemented using one or more integrated circuit devices, application-specific integrated circuits (ASICs), and / or memory devices. For example, the RISC core can include an instruction cache and / or tightly coupled RAM.

[0087] The DMA may enable the PVA component to access system memory independently of the one or more CPUs 606. The DMA may support any functionality for optimizing the PVA, including, but not limited to, supporting multi-dimensional addressing and / or circular addressing. In some examples, the DMA may support addressing in up to six or more dimensions, which may include block width, block height, block depth, horizontal block stride, vertical block stride, and / or depth stride.

[0088] A vector processor can be a programmable processor designed to efficiently and flexibly execute programming of computer vision algorithms and provide signal processing capabilities. In some examples, a PVA may include a PVA core and two vector processing subsystem partitions. The PVA core may include a processor subsystem, one or more DMA engines (e.g., two DMA engines), and / or other peripherals. The vector processing subsystem may operate as the main processing engine of the PVA and may include a vector processing unit (VPU), an instruction cache, and / or a vector memory (e.g., VMEM). The VPU core may include a digital signal processor, such as a single instruction, multiple data (SIMD), a very long instruction word (VLIW) digital signal processor. The combination of SIMD and VLIW can increase throughput and speed.

[0089] Each vector processor may include an instruction cache and may be coupled to dedicated memory. As a result, in some examples, each vector processor may be configured to execute independently of the other vector processors. In other examples, the vector processors included in a particular PVA may be configured to employ data parallelism. For example, in some embodiments, multiple vector processors included in a single PVA may execute the same computer vision algorithm on different regions of an image. In other examples, the vector processors included in a particular PVA may execute different computer vision algorithms simultaneously on the same image, or even execute different algorithms on sequenced images or portions of images. Among other things, any number of PVAs may be included in a hardware acceleration cluster, and any number of vector processors may be included in each PVA. In addition, the PVAs may include additional error correction code (ECC) memory to enhance overall system security.

[0090] One or more accelerators 614 (e.g., a hardware acceleration cluster) may include a computer vision network on-chip and SRAM to provide high bandwidth, low latency SRAM to one or more accelerators 614. In some examples, the on-chip memory may include at least 4MB of SRAM, including, but not limited to, eight field-configurable memory blocks accessible by the PVA and DLA. Each pair of memory blocks may include an advanced peripheral bus (APB) interface, configuration circuitry, a controller, and a multiplexer. Any type of memory may be used. The PVA and DLA may access the memory via a backbone that provides high-speed memory access to the PVA and DLA. The backbone may include an on-chip computer vision network that interconnects the PVA and DLA to the memory (e.g., using APB).

[0091] The on-chip computer vision network can include an interface that ensures that both the PVA and DLA provide ready and valid signals before transmitting any control signals, addresses, or data. This interface can provide independent phases and independent channels for transmitting control signals, addresses, and data, as well as burst-based communication for continuous data transmission. This interface can comply with ISO 26262 or IEC 61508, although other standards and protocols may also be used.

[0092] In some examples, one or more SoCs 604 may include a real-time ray tracing hardware accelerator, as described in U.S. Patent Application No. 16 / 101,232, filed on August 10, 2018. The real-time ray tracing hardware accelerator may be used to quickly and efficiently determine the position and range of objects (e.g., within a world model), generate real-time visual simulations for radar signal interpretation, sound propagation synthesis and / or analysis, sonar system simulations, general wave propagation simulations, comparison with lidar data for positioning, and / or other functions and / or other uses. In some embodiments, one or more tree traversal units (TTUs) may be used to perform one or more ray tracing related operations.

[0093] One or more accelerators 614 (e.g., a hardware accelerator cluster) have a wide range of uses in autonomous driving. The PVA may be a programmable vision accelerator that can be used in key processing stages of ADA and autonomous vehicles. The capabilities of the PVA are well suited for algorithmic domains that require predictable processing with low power consumption and low latency. In other words, the PVA performs well on semi-intensive or regular intensive computations, even on small data sets, which require predictable runtimes with low latency and low power consumption. Therefore, in the context of autonomous vehicle platforms, the PVA is designed to run classic computer vision algorithms because they are very efficient at object detection and integer math operations.

[0094] For example, according to one embodiment of the present technology, the PVA is used to perform computer stereo vision. In some examples, a semi-global matching-based algorithm can be used, although this is not intended to be limiting. Many applications for Level 3-5 autonomous driving require on-the-fly motion estimation / stereo matching (e.g., structure from motion, pedestrian recognition, lane detection, etc.). The PVA can perform computer stereo vision functions on input from two monocular cameras.

[0095] In some examples, PVA can be used to perform dense optical flow, providing processed radar data based on processing raw radar data (e.g., using a 4D Fast Fourier Transform). In other examples, PVA is used for time-of-flight depth processing, e.g., by processing raw time-of-flight data to provide processed time-of-flight data.

[0096] The DLA can be used to run any type of network to enhance control and driving safety, including, for example, a neural network that outputs a confidence measure for each object detection. Such confidence values ​​can be interpreted as probabilities, or provide a relative "weight" of each detection relative to other detections. This confidence value enables the system to further determine which detections should be considered true positives versus false positives. For example, the system can set a confidence threshold and only consider detections exceeding the threshold as true positives. In an automatic emergency braking (AEB) system, false positive detections would cause the vehicle to automatically apply emergency braking, which is clearly undesirable. Therefore, only the most reliable detections should be considered triggers for AEB. The DLA can run a neural network to regress the confidence value. The neural network can take as input at least a subset of parameters, such as bounding box dimensions, a ground plane estimate obtained (e.g., from another subsystem), inertial measurement unit (IMU) sensor 666 output related to the vehicle 600's orientation and range, and three-dimensional position estimates of objects obtained from the neural network and / or other sensors (e.g., lidar sensor 664 or radar sensor 660).

[0097] One or more SoCs 604 may include one or more data stores 616 (e.g., memory). Data store 616 may be on-chip memory of SoC 604 that may store neural networks to be executed on the GPU and / or DLA. In some examples, the capacity of data store 616 may be large enough to store multiple neural network instances for redundancy and safety. Data store 612 may include a L2 or L3 cache 612. As described herein, references to one or more data stores 616 may include references to memory associated with the PVA, DLA, and / or one or more other accelerators 614.

[0098] One or more SoCs 604 may include one or more processors 610 (e.g., embedded processors). Processors 610 may include a boot and power management processor, which may be a dedicated processor and subsystem for handling boot power and management functions and related security implementations. The boot and power management processor may be part of the boot sequence of one or more SoCs 604 and may provide runtime power management services. The boot power and management processor may provide clock and voltage programming, assist with system low power state transitions, manage SoC 604 thermal and temperature sensors, and / or manage SoC 604 power states. Each temperature sensor may be implemented as a ring oscillator with an output frequency proportional to temperature, and one or more SoCs 604 may use the ring oscillator to detect the temperature of one or more CPUs 606, one or more GPUs 608, and / or one or more accelerators 614. If the temperature is determined to exceed a threshold, the boot and power management processor may enter a temperature fault routine and place one or more SoCs 604 in a low power state and / or place the vehicle 600 in a driver safety stop mode (e.g., to safely stop the vehicle 600).

[0099] The one or more processors 610 may also include a set of embedded processors that can serve as an audio processing engine. The audio processing engine can be an audio subsystem that is capable of full hardware support for multi-channel audio through multiple interfaces and a wide and flexible audio I / O interface. In some examples, the audio processing engine is a dedicated processor core of a digital signal processor with dedicated RAM.

[0100] One or more processors 610 may also include an always-on processor engine that provides the necessary hardware functionality to support low-power sensor management and wake-up use cases. The always-on processor engine may include a processor core, tightly coupled RAM, supporting peripherals (such as timers and interrupt controllers), various I / O controller peripherals, and routing logic.

[0101] One or more processors 610 may also include a safety cluster engine, which includes a dedicated processor subsystem for handling safety management for automotive applications. The safety cluster engine may include two or more processor cores, tightly coupled RAM, supporting peripherals (e.g., timers, interrupt controllers, etc.), and / or routing logic. In safety mode, the two or more cores can operate in lockstep mode and act as a single core with comparison logic to detect any differences between their operations.

[0102] The one or more processors 610 may also include a real-time camera engine, which may include a dedicated processor subsystem for handling real-time camera management.

[0103] The one or more processors 610 can also include a high dynamic range signal processor, which can include an image signal processor that is a hardware engine that is part of the camera processing pipeline.

[0104] The one or more processors 610 can include a video image compositor, which can 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 player window. The video image compositor can perform lens distortion correction on the one or more wide-angle cameras 670, the one or more surround cameras 674, and / or the in-cabin monitoring camera sensors. The in-cabin monitoring camera sensors are preferably monitored by a neural network running on another instance of the advanced SoC that is configured to identify in-cabin events and respond accordingly. The in-cabin system can perform lip reading to activate cellular services and place a phone call, dictate an email, change the vehicle destination, activate or change the vehicle’s infotainment system and settings, or provide voice-activated web surfing. Certain functions can only be available to the driver when the vehicle is operating in an autonomous mode, otherwise they will be disabled.

[0105] The video image compositor can include enhanced temporal noise reduction for spatial and temporal noise reduction. For example, where motion occurs in a video, the noise reduction appropriately weights the spatial information, reducing the weight of information provided by adjacent frames. Where an image or portion of an image does not include motion, the temporal noise reduction performed by the video image compositor can use information from a previous image to reduce noise in the current image.

[0106] The video image compositor can also be configured to perform stereo correction on input stereoscopic lens frames. The video image compositor can also be used for user interface composition when the operating system desktop is in use, and does not require the GPU 608 to continuously render new surfaces. The video image compositor can be used to offload the one or more GPUs 608 to improve performance and responsiveness, even when the one or more GPUs 608 are powered on and active when performing 3D rendering.

[0107] The one or more SoCs 604 can also include a Mobile Industry Processor Interface (MIPI) camera serial interface for receiving video and input from cameras, a high-speed interface, and / or a video input block that can be used for camera and related pixel input functions. The one or more SoCs 604 can also include one or more input / output controllers that can be controlled by software and can be used to receive I / O signals that do not have a specific role assigned.

[0108] The one or more SoCs 604 may also include a wide range of peripheral interfaces to enable communication with peripheral devices, audio codecs, power management, and / or other devices. The one or more SoCs 604 may be used to process data from cameras (e.g., via Gigabit multimedia serial links and Ethernet connections), sensors (e.g., one or more lidar sensors 664, one or more radar sensors 660, etc., which may be connected via Ethernet), data from the bus 602 (e.g., vehicle 600 speed, steering wheel position, etc.), data from one or more GNSS sensors 658 (e.g., connected via Ethernet or CAN bus). The one or more SoCs 604 may also include a dedicated high-performance mass storage controller, which may include its own DMA engine and may be used to free the one or more CPUs 606 from routine data management tasks.

[0109] One or more SoCs 604 can be an end-to-end platform with a flexible architecture that spans automation levels 3-5, thereby providing a comprehensive functional safety architecture that leverages and effectively uses computer vision and ADAS technologies to achieve diversity and redundancy, providing a platform for flexible and reliable driver software stacks and deep learning tools. Compared to traditional systems, one or more SoCs 604 can be faster, more reliable, and even more energy-efficient and space-efficient. For example, when one or more accelerators 614 are combined with one or more CPUs 606, one or more GPUs 608, and one or more data stores 616, a fast and efficient platform can be provided for Level 3-5 autonomous vehicles.

[0110] As a result, this technology offers capabilities and functionality not possible with traditional 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 often fail 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, a requirement for in-vehicle ADAS applications and for practical Level 3-5 autonomous vehicles.

[0111] Compared to traditional systems, by providing a CPU complex, a GPU complex, and a hardware acceleration cluster, the technology described herein allows for the simultaneous and / or sequential execution of multiple neural networks and allows the results to be combined to achieve Level 3-5 autonomous driving capabilities. For example, a CNN executed on a DLA or dGPU (e.g., one or more GPUs 620) can 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 can also include a neural network that can recognize, interpret, and provide semantic understanding of the signs and pass this semantic understanding to a path planning module running on the CPU complex.

[0112] As another example, multiple neural networks can be run simultaneously, as required for Level 3, 4, or 5 driving. For example, a warning sign consisting of "Warning: Flashing lights indicate icing conditions" and a light can be interpreted by multiple neural networks independently or collectively. The sign itself can be recognized as a traffic sign by a first deployed neural network (e.g., a trained neural network), and the text "Flashing lights indicate icing conditions" can be interpreted by a second deployed neural network, which, when the flashing lights are detected, notifies the vehicle's routing software (preferably executing on a CPU complex) of the presence of icing conditions. A third deployed neural network can be operated over multiple frames to identify the flashing lights, notifying the vehicle's routing software of the presence of the flashing lights. All three neural networks can run simultaneously, for example, within the DLA and / or on one or more GPUs 608.

[0113] In some examples, a CNN for facial recognition and owner recognition can use data from camera sensors to identify the presence of an authorized driver and / or owner of vehicle 600. The always-on sensor processing engine can be used to unlock the vehicle when the owner approaches the driver's door and turns on the lights, and, in security mode, disable the vehicle when the owner leaves the vehicle. In this way, one or more SoCs 604 provide protection against theft and / or carjacking.

[0114] In another example, a CNN for emergency vehicle detection and identification can use data from microphone 696 to detect and identify emergency vehicle sirens. Unlike traditional systems that use general classifiers to detect sirens and manually extract features, one or more SoC(s) 604 use CNNs to classify environmental and urban sounds, as well as classify visual data. In a preferred embodiment, the CNN running on the DLA is trained to identify the relative closing speeds of emergency vehicles (e.g., by using the Doppler effect). The CNN can also be trained to identify emergency vehicles specific to the area in which the vehicle is operating, as identified by one or more GNSS sensors 658. Thus, for example, when operating in Europe, the CNN will seek to detect European sirens, and when in the United States, the CNN will seek to identify only North American sirens. Once an emergency vehicle is detected, a control program can be used with the help of ultrasonic sensors 662 to execute emergency vehicle safety routines to slow the vehicle, pull over, stop the vehicle, and / or idle the vehicle until one or more emergency vehicles pass.

[0115] The vehicle may include one or more CPUs 618 (e.g., one or more discrete CPUs or one or more dCPUs) that may be coupled to one or more SoCs 604 via a high-speed interconnect (e.g., PCIE). For example, one or more CPUs 618 may include an X86 processor. The CPUs 618 may be used to perform any of a variety of functions, including arbitrating potentially inconsistent results between ADAS sensors and the SoC 604, and / or monitoring the status and health of one or more controllers 636 and / or the infotainment SoC 630.

[0116] The vehicle 600 may include one or more GPUs 620 (e.g., one or more discrete GPUs or one or more dGPUs) that may be coupled to the SoC 604 via a high-speed interconnect (e.g., NVIDIA's NVLINK). The one or more GPUs 620 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 the neural network based on input (e.g., sensor data) from sensors of the vehicle 600.

[0117] The vehicle 600 may also include a network interface 624, which may include one or more wireless antennas 626 (e.g., one or more wireless antennas for different communication protocols, such as a cellular antenna, a Bluetooth antenna, etc.). The network interface 624 may be used to enable wireless connections to the cloud (e.g., to one or more servers 678 and / or other network devices), to other vehicles, and / or to computing devices (e.g., a passenger's client device) via the Internet. To communicate with other vehicles, a direct link may be established between the two vehicles and / or an indirect link may be established (e.g., via a network and the Internet). A vehicle-to-vehicle communication link may be used to provide a direct link. The vehicle-to-vehicle communication link may provide the vehicle 600 with information about vehicles in the vicinity of the vehicle 600 (e.g., vehicles in front of, to the sides of, and / or behind the vehicle 600). This functionality may be part of the cooperative adaptive cruise control functionality of the vehicle 600.

[0118] The network interface 624 may include a SoC that provides modulation and demodulation functionality and enables one or more controllers 636 to communicate over a wireless network. The network interface 624 may include an RF front-end for up-conversion from baseband to RF and down-conversion from RF to baseband. The frequency conversion may be performed by well-known processes and / or may be performed using a superheterodyne process. In some examples, the RF front-end functionality may be provided by a separate chip. The network interface may include wireless functionality for communicating via LTE, WCDMA, UMTS, GSM, CDMA2000, Bluetooth, Bluetooth LE, Wi-Fi, Z-Wave, ZigBee, LoRaWAN, and / or other wireless protocols.

[0119] The vehicle 600 may further include one or more data stores 628, which may include off-chip (e.g., off-SoC). The data stores 628 may include one or more storage elements, including RAM, SRAM, DRAM, VRAM, flash memory, a hard disk, and / or other components and / or devices capable of storing at least one bit of data.

[0120] The vehicle 600 may also include one or more GNSS sensors 658. One or more GNSS sensors 658 (e.g., GPS, assisted GPS sensors, differential GPS (DGPS) sensors, etc.) are used to assist with mapping, perception, occupancy grid generation, and / or path planning functions. Any number of GNSS sensors 658 may be used, including, for example, but not limited to, GPS using a USB connector with an Ethernet to serial (RS-232) bridge.

[0121] Vehicle 600 may also include one or more radar sensors 660. Vehicle 600 may use one or more radar sensors 660 for remote vehicle detection, even in darkness and / or inclement weather conditions. The radar functional safety level may be ASIL B. One or more radar sensors 660 may use CAN and / or bus 602 (e.g., to transmit data generated by one or more radar sensors 660) for control and access to target tracking data, and in some examples, access to raw data via Ethernet. A variety of radar sensor types may be used. For example, but not limited to, one or more radar sensors 660 may be suitable for front, rear, and side radar use. In some examples, a pulse Doppler radar sensor is used.

[0122] One or more radar sensors 660 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 can be used for adaptive cruise control functions. Long-range radar systems can provide a wide field of view achieved through two or more independent scans, for example, within a range of 250 meters. One or more radar sensors 660 can help distinguish between static and moving objects and can be used by ADAS systems for emergency brake assistance and forward collision warning. Long-range radar sensors may include monostatic multimode radars with multiple (e.g., six or more) fixed radar antennas and high-speed CAN and FlexRay interfaces. In an example with six antennas, the central four antennas can create a focused beam pattern designed to record the 600 environment around the vehicle at higher speeds with minimal interference from traffic in adjacent lanes. The additional two antennas can expand the field of view, making it possible to quickly detect vehicles entering or leaving the vehicle's 600 lane.

[0123] For example, a medium-range radar system may include a range of up to 660 meters (front) or 80 meters (rear), and a field of view of up to 42 degrees (front) or 650 degrees (rear). Short-range radar systems may include, but are not limited to, radar sensors designed to be mounted on either end of the rear bumper. When mounted on either end of the rear bumper, such radar sensor systems may generate two beams to continuously monitor the blind spots behind and beside the vehicle.

[0124] Short-range radar systems may be used in ADAS systems for blind spot detection and / or lane change assistance.

[0125] The vehicle 600 can also include one or more ultrasonic sensors 662. The one or more ultrasonic sensors 662 can be located at the front, rear, and / or sides of the vehicle 600, can be used for parking assistance and / or to create and update the occupancy grid. A variety of ultrasonic sensors 662 can be used, with different ultrasonic sensors 662 being used for different detection ranges (e.g., 2.5m, 4m). The one or more ultrasonic sensors 662 can operate at an ASIL B level of functional safety.

[0126] The vehicle 600 can include one or more lidar sensors 664. The one or more lidar sensors 664 can be used for object and pedestrian detection, emergency braking, collision avoidance, and / or other functions. The functional safety rating of the one or more lidar sensors 664 can be ASIL B. In some examples, the vehicle 600 can include multiple lidar sensors 664 (e.g., two, four, six, etc.) that can use Ethernet (e.g., to provide data to a Gigabit Ethernet switch).

[0127] In some examples, the one or more lidar sensors 664 can be capable of providing a list of objects and their distances in a 360-degree field of view. The advertised range of one or more commercial lidar sensors 664 can be approximately 600m, with an accuracy of 2cm-3cm, and support, for example, 600Mbps Ethernet connections. In some examples, one or more flush lidar sensors 664 can be used. In such examples, the one or more lidar sensors 664 can be implemented as small devices that can be embedded in the front, rear, sides, and / or corners of the vehicle 600. In such examples, the one or more lidar sensors 664 can provide up to 120 degrees of horizontal and 35 degrees of vertical field of view, with a range of 200m even for low reflectivity objects. Front-mounted one or more lidar sensors 664 can be configured for a horizontal field of view between 45 degrees and 135 degrees.

[0128] In some examples, lidar technology, such as 3D flash lidar, may also be used. 3D flash lidar uses a laser flash as a transmission source, illuminating approximately 200 meters around the vehicle. The flash lidar device includes a receiver that records the laser pulse transmission time and the reflected light at each pixel, where each pixel corresponds to the range from the vehicle to the object. Flash lidar can use each laser flash to generate a highly accurate, distortion-free image of the surroundings. In some examples, four flash lidar sensors may be deployed, one on each side of vehicle 600. Available 3D flash lidar systems include solid-state 3D staring array lidar cameras with no moving parts other than the fan (e.g., non-scanning lidar devices). Flash lidar devices can use 5 nanosecond Class I (eye-safe) laser pulses per frame and capture the reflected laser light in the form of a 3D range point cloud and co-registered intensity data. By using flash lidar, and because flash lidar is a solid-state device with no moving parts, one or more lidar sensors 664 can be less susceptible to motion blur, vibration, and / or shock.

[0129] The vehicle may also include one or more IMU sensors 666. In some examples, the one or more IMU sensors 666 may be located in the center of the rear axle of the vehicle 600. The one or more IMU sensors 666 may include, for example, but not limited to, one or more accelerometers, one or more magnetometers, one or more gyroscopes, one or more magnetic compasses, and / or other sensor types. In some examples, such as in a six-axis application, the one or more IMU sensors 666 may include an accelerometer and a gyroscope, while in a nine-axis application, the one or more IMU sensors 666 may include an accelerometer, a gyroscope, and a magnetometer.

[0130] In some embodiments, one or more IMU sensors 666 may be implemented as a miniature, high-performance GPS-assisted inertial navigation system (GPS / INS) that combines micro-electromechanical system (MEMS) inertial sensors, a high-sensitivity GPS receiver, and advanced Kalman filtering algorithms to provide estimates of position, velocity, and attitude. Thus, in some examples, one or more IMU sensors 666 may enable the vehicle 600 to estimate heading without input from a magnetic sensor by directly observing and correlating velocity changes from GPS to one or more IMU sensors 666. In some examples, one or more IMU sensors 666 and one or more GNSS sensors 658 may be combined in a single integrated unit.

[0131] The vehicle may include one or more microphones 696 positioned within and / or around the vehicle 600. The one or more microphones 696 may be used for emergency vehicle detection and identification, among other things.

[0132] The vehicle may also include any number of camera types, including one or more stereo cameras 668, one or more wide angle cameras 670, one or more infrared cameras 672, one or more surround cameras 674, one or more long-range and / or mid-range cameras 698, and / or other camera types. The cameras may be used to capture image data of the entire periphery of the vehicle 600. The type of camera used depends on the embodiment and requirements of the vehicle 600, and any combination of camera types may be used to provide the necessary coverage around the vehicle 600. Furthermore, 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 other numbers of cameras. As examples, the cameras may support, but are not limited to, Gigabit Multimedia Serial Link (GMSL) and / or Gigabit Ethernet. This document will refer to Figure 6A and Figure 6B Describe each camera in more detail.

[0133] Vehicle 600 may also include one or more vibration sensors 642. One or more vibration sensors 642 can measure vibrations of vehicle components, such as axles. For example, changes in vibration may indicate changes in the road surface. In another example, when two or more vibration sensors 642 are used, the difference in vibrations can be used to determine friction or slip in the road surface (e.g., when there is a difference in vibration between a powered axle and a freely rotating axle).

[0134] Vehicle 600 may include an ADAS system 638. In some examples, ADAS system 638 may include a SoC. ADAS system 638 may include automatic / adaptive / auto cruise control (ACC), cooperative adaptive cruise control (CACC), forward collision warning (FCW), automatic emergency braking (AEB), lane departure warning (LDW), lane keeping assist (LKA), blind spot warning (BSW), rear cross traffic alert (RCTW), collision warning system (CWS), lane centering (LC), and / or other features and functions.

[0135] The ACC system may utilize one or more radar sensors 660, one or more lidar sensors 664, and / or one or more cameras. The ACC system may include longitudinal ACC and / or lateral ACC. Longitudinal ACC monitors and controls the distance to the vehicle directly in front of vehicle 600 and automatically adjusts vehicle speed to maintain a safe distance from the vehicle in front. Lateral ACC maintains distance and recommends lane changes when necessary. Lateral ACC is associated with other ADAS applications, such as LCA and CWS.

[0136] CACC uses information from other vehicles, which may be received from other vehicles indirectly via a network interface 624 and / or one or more wireless antennas 626, via a wireless link, or via a network connection (e.g., via the Internet). A direct link may be provided by a vehicle-to-vehicle (V2V) communication link, while an indirect link may be an infrastructure-to-vehicle (I2V) communication link. Generally speaking, the V2V communication concept provides information about the vehicle in front (e.g., the vehicle directly in front of the vehicle 600 and in the same lane), while the I2V communication concept provides information about the traffic ahead. A CACC system may include one or both of the I2V and V2V information sources. Taking into account the information about the vehicle in front of the vehicle 600, CACC may be more reliable and may potentially improve the smoothness of traffic flow and reduce congestion on the road.

[0137] The FCW system is designed to alert the driver to hazards so that the driver can take corrective action. The FCW system uses a forward-facing camera and / or one or more radar sensors 660 coupled to a dedicated processor, DSP, FPGA, and / or ASIC, which is electrically coupled to driver feedback, such as a display, speaker, and / or vibration component. The FCW system can provide warnings such as audible and visual warnings, vibrations, and / or rapid brake pulses.

[0138] The AEB system detects an imminent forward collision with another vehicle or other object and may automatically apply the brakes if the driver does not take corrective action within specified time or distance parameters. The AEB system may utilize one or more forward-facing cameras and / or one or more radar sensors 660 coupled to a dedicated processor, DSP, FPGA, and / or ASIC. When the AEB system detects a hazard, it typically first alerts the driver to take corrective action to avoid the collision. If the driver does not take corrective action, the AEB system may automatically apply the brakes to prevent or at least mitigate the effects of the anticipated collision. The AEB system may include technologies such as dynamic brake support and / or collision approach braking.

[0139] The LDW system provides visual, audible, and / or tactile warnings, such as steering wheel or seat vibrations, to alert the driver when the vehicle 600 crosses a lane marking. The LDW system does not activate when the driver indicates an intention to leave the lane by activating a turn signal. The LDW system may utilize a front-facing camera coupled to a dedicated processor, DSP, FPGA, and / or ASIC that is electrically coupled to driver feedback, such as a display, speaker, and / or vibration component.

[0140] The LKA system is a variation of the LDW system. If the vehicle 600 begins to leave its lane, the LKA system provides steering input or braking to correct the vehicle 600.

[0141] The BSW system detects and warns the driver of vehicles in the vehicle's blind spot. The BSW system can provide visual, audible, and / or tactile alerts to indicate that merging or changing lanes is unsafe. The system may provide additional warnings when the driver uses the turn signal. The BSW system can use a rear-facing camera and / or radar sensor 660 coupled to a dedicated processor, DSP, FPGA, and / or ASIC that is electrically coupled to driver feedback, such as a display, speaker, and / or vibration component.

[0142] When the vehicle 600 detects an object outside the range of the rear camera while backing up, the RCTW system can provide visual, audible, and / or tactile notifications. Some RCTW systems include automatic emergency braking (AEB) to ensure that the vehicle brakes are applied to avoid a collision. The RCTW system can use one or more rear-facing radar sensors 660 coupled to a dedicated processor, DSP, FPGA, and / or ASIC that is electrically coupled to driver feedback, such as a display, speaker, and / or vibration assembly.

[0143] Conventional ADAS systems can be prone to false positive results, which can be irritating and distracting to the driver, but typically do not have catastrophic consequences because the ADAS system alerts the driver and allows the driver to decide whether a safe condition actually exists and take appropriate action. However, in the autonomous vehicle 600, in the event of conflicting results, the vehicle 600 itself must decide whether to listen to the results from the primary computer or the secondary computer (e.g., the first controller 636 or the second controller 636). For example, in some embodiments, the ADAS system 638 can be a backup and / or secondary computer that provides perception information to the backup computer module. The backup computer rationality monitor can run redundant, different software on hardware components to detect failures in perception and dynamic driving tasks. The output from the ADAS system 638 can be provided to the monitoring MCU. If the outputs of the primary and secondary computers conflict, the monitoring MCU must determine how to reconcile the conflict to ensure safe operation.

[0144] In some examples, the primary computer can be configured to provide a confidence score to the monitoring MCU, indicating the primary computer's confidence in the selected result. If the confidence score exceeds a threshold, the monitoring MCU can follow the primary computer's direction, regardless of whether the secondary computer provides conflicting or inconsistent results. If the confidence score does not meet the threshold and the primary and secondary computers indicate different results (e.g., conflicting), the monitoring MCU can arbitrate between the computers to determine the appropriate result.

[0145] The monitoring MCU can be configured to run one or more neural networks that have been trained and configured to determine, based on the outputs of the primary and secondary computers, the conditions under which the secondary computer provides a false alarm. Thus, the one or more neural networks in the monitoring MCU can learn when the outputs of the secondary computer are trustworthy and when they are not. For example, when the secondary computer is a radar-based FCW system, the one or more neural networks in the monitoring MCU can learn when the FCW system identifies a metal object that is not actually a danger, such as a drain grate or manhole cover, that triggers an alarm. Similarly, when the secondary computer is a camera-based LDW system, the neural network in the monitoring MCU can learn to override the LDW when a bicycle or pedestrian is present and lane departure is actually the safest maneuver. In embodiments that include one or more neural networks running on the monitoring MCU, the monitoring MCU can include at least one of a DLA or GPU adapted to run the one or more neural networks with associated memory. In a preferred embodiment, the monitoring MCU can include and / or be a component of the SoC 604.

[0146] In other examples, the ADAS system 638 may include an auxiliary computer that uses traditional computer vision rules to perform ADAS functions. Thus, the auxiliary computer may use classic computer vision rules (if-then), and the presence of one or more neural networks in the monitoring MCU may improve reliability, safety, and performance. For example, the diverse implementations and intentional non-identification make the entire system more fault-tolerant, particularly to failures caused by software (or software-hardware interface) functions. For example, if there is a software defect or error in the software running on the main computer, and a different software code running on the auxiliary computer provides the same overall result, the monitoring MCU may have greater confidence that the overall result is correct and that the software or hardware defect on the main computer did not cause a significant error.

[0147] In some examples, the output of ADAS system 638 can be fed into the primary computer's perception block and / or the primary computer's dynamic driving task block. For example, if ADAS system 638 indicates a forward collision warning due to an object directly ahead, the perception block can use this information when identifying the object. In other examples, as described herein, the secondary computer can have its own trained neural network, thereby reducing the risk of false positives.

[0148] The vehicle 600 may also include an infotainment SoC 630 (e.g., an in-vehicle infotainment system (IVI)). Although shown and described as an SoC, the infotainment system may not be a SoC and may include two or more discrete components. The infotainment SoC 630 may include a combination of hardware and software that can be used to provide audio (e.g., music, personal digital assistant, navigation instructions, news, radio, etc.), video (e.g., TV, movies, streaming, etc.), phone (e.g., hands-free calling), network connectivity (e.g., LTE, Wi-Fi, etc.), and / or provide information services to the vehicle 600 (e.g., a navigation system, rear parking assist, radio data system, vehicle-related information such as fuel level, total distance traveled, brake fuel level, engine oil level, door open / closed, air filter information, etc.). For example, the infotainment SoC 630 may be a radio, a disk player, a navigation system, a video player, USB and Bluetooth connectivity, an onboard computer, in-car entertainment, Wi-Fi, steering wheel audio controls, hands-free voice control, a heads-up display (HUD), an HMI display 634, 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 630 may also be used to provide information (e.g., visual and / or auditory) to a vehicle user, such as information from an ADAS system 638, autonomous driving information, such as planned vehicle maneuvers, trajectories, surrounding environment information (e.g., intersection information, vehicle information, road information, etc.), and / or other information.

[0149] The infotainment SoC 630 may include GPU functionality. The infotainment SoC 630 may communicate with other devices, systems, and / or components of the vehicle 600 via a bus 602 (e.g., a CAN bus, Ethernet, etc.). In some examples, the infotainment SoC 630 may be coupled to a supervisory MCU so that the GPU of the infotainment system can perform some self-driving functions in the event of a failure of one or more main controllers 636 (e.g., the main computer and / or backup computer of the vehicle 600). In such an example, the infotainment SoC 630 may place the vehicle 600 in a driver-to-safety parking mode, as described herein.

[0150] The vehicle 600 may also include an instrument cluster 632 (e.g., a digital instrument panel, an electronic instrument cluster, a digital instrument panel, etc.). The instrument cluster 632 may include a controller and / or a supercomputer (e.g., a discrete controller or a supercomputer). The instrument cluster 632 may include a set of instruments, such as a speedometer, fuel level, oil pressure, a tachometer, an odometer, turn indicators, a shift position indicator light, one or more seat belt warning lights, one or more parking brake warning lights, one or more engine check lights, 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 630 and the instrument cluster 632. In other words, the instrument cluster 632 may be included as part of the infotainment SoC 630, and vice versa.

[0151] Figure 6D One or more cloud-based servers according to some embodiments of the present disclosure and Figure 6A 6. System diagram of communication between an exemplary autonomous vehicle 600. System 676 may include one or more servers 678, one or more networks 690, and vehicles, including vehicle 600. One or more servers 678 may include multiple GPUs 684(a)-684(H) (collectively, GPUs 684), PCIe switches 682(a)-682(H) (collectively, PCIe switches 682), and / or CPUs 680(A)-680(B) (collectively, herein, CPUs 680). GPUs 684, CPUs 680, and PCIe switches may be interconnected via a high-speed interconnect, such as, but not limited to, NVLink interface 688 and / or PCIE connection 686 developed by NVIDIA. In some examples, GPUs 684 are connected via NVLink and / or NV switch SoCs, and GPUs 684 and PCIe switches 682 are connected via a PCIe interconnect. Although eight GPUs 684, two CPUs 680, and two PCIe switches are illustrated, this is not intended to be limiting. Depending on the embodiment, each of the one or more servers 678 may include any number of GPUs 684, CPUs 680, and / or PCIe switches. For example, one or more servers 678 may each include eight, sixteen, thirty-two, and / or more GPUs 684.

[0152] One or more servers 678 can receive image data representing an image from a vehicle via one or more networks 690 that shows unexpected or changing road conditions, such as recently initiated roadwork. One or more servers 678 can send neural networks 692, updated neural networks 692, and / or map information 694 to the vehicle via one or more networks 690, including information about traffic and road conditions. Updates to map information 694 can include updates to HD map 622, such as information about construction sites, potholes, detours, flooding, and / or other obstacles. In some examples, neural networks 692, updated neural networks 692, and / or map information 694 can be derived from new training and / or experience represented by data received from any number of vehicles in the environment, and / or based on training performed at a data center (e.g., using one or more servers 678 and / or other servers).

[0153] One or more servers 678 can be used to train a machine learning model (e.g., a neural network) based on training data. The training data can be generated by the vehicle and / or can be generated in simulation (e.g., using a game engine). In some examples, the training data is labeled (e.g., neural networks benefit from supervised learning) and / or undergoes other preprocessing, while in other examples, the training data is not labeled and / or preprocessed (e.g., neural networks do not require supervised learning). Training can 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 alternative dictionary learning), rule-based machine learning, anomaly detection, and any variants or combinations thereof. Once the machine learning model is trained, the machine learning model can be used by the vehicle (e.g., transmitted to the vehicle 690 via one or more networks, and / or the machine learning model can be used by one or more servers 678 to remotely monitor the vehicle).

[0154] In some examples, one or more servers 678 can receive data from the vehicle and apply the data to the latest real-time neural network to perform real-time intelligent reasoning. One or more servers 678 may include deep learning supercomputers and / or dedicated AI computers powered by GPUs 684, such as DGX and DGX Station machines developed by NVIDIA. However, in some examples, one or more servers 678 may include the deep learning infrastructure of a data center that uses only CPU power.

[0155] The deep learning infrastructure of one or more servers 678 can perform rapid real-time inference and can use this capability to assess and verify the health of the processor, software, and / or associated hardware in the vehicle 600. For example, the deep learning infrastructure can receive periodic updates from the vehicle 600, such as a sequence of images and / or objects that the vehicle 600 has located within that sequence of images (e.g., via computer vision and / or other machine learning object classification techniques). The deep learning infrastructure can run its own neural network to identify objects and compare them to the objects identified by the vehicle 600. If the results do not match and the infrastructure concludes that the AI ​​in the vehicle 600 has failed, the one or more servers 678 can send a signal to the vehicle 600, instructing the vehicle's 600 fail-safe computer to assume control, notify passengers, and complete a safe parking maneuver.

[0156] For inference, one or more servers 678 may include one or more GPUs 684 and one or more programmable inference accelerators (e.g., NVIDIA's TensorRT). The combination of GPU-powered servers and inference acceleration may enable real-time responses. In other examples, such as where performance is less critical, servers powered by CPUs, FPGAs, and other processors may be used for inference.

[0157] Example computing device

[0158] Figure 7 7 is a block diagram of an example computing device 700 suitable for implementing some embodiments of the present disclosure. The computing device 700 may include an interconnect system 702 that directly or indirectly couples the following devices: memory 704, one or more central processing units (CPUs) 706, one or more graphics processing units (GPUs) 708, a communication interface 710, input / output (I / O) ports 712, input / output components 714, a power supply 716, one or more presentation components 718 (e.g., a display), and one or more logic units 720. In at least one embodiment, the one or more computing devices 700 may include one or more virtual machines (VMs), and / or any of their components may include virtual components (e.g., virtual hardware components). For non-limiting examples, the one or more GPUs 708 may include one or more vGPUs, the one or more CPUs 706 may include one or more vCPUs, and / or the one or more logic units 720 may include one or more virtual logic units. Thus, one or more computing devices 700 may include discrete components (eg, a complete GPU dedicated to computing device 700 ), virtual components (eg, a portion of a GPU dedicated to computing device 700 ), or a combination thereof.

[0159] although Figure 7The various modules in FIG70 are shown as being connected to each other via interconnect system 702, but this is not intended to be limiting and is provided for clarity. For example, in some embodiments, presentation components 718, such as a display device, may be considered I / O components 714 (e.g., if the display is a touch screen). As another example, CPU 706 and / or GPU 708 may include memory (e.g., memory 704 may represent a storage device in addition to the memory of GPU 708, CPU 706, and / or other components). In other words, Figure 7 The computing devices referred to in the present disclosure are illustrative only. No distinction is made between "workstations," "servers," "laptops," "desktops," "tablets," "client devices," "mobile devices," "handheld devices," "game consoles," "electronic control units (ECUs)," "virtual reality systems," and / or other device or system types, as in Figure 7 as contemplated within the scope of computing devices.

[0160] The interconnection system 702 may represent one or more links or buses, such as an address bus, a data bus, a control bus, or a combination thereof. The interconnection system 702 may include one or more bus or link 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 Direct (PCIE) bus, and / or other types of buses or links. In some embodiments, there is a direct connection between components. For example, the CPU 706 may be directly connected to the memory 704. In addition, the CPU 706 may be directly connected to the GPU 708. In the case where there is a direct connection or a point-to-point connection between components, the interconnection system 702 may include a PCIe link for performing the connection. In these examples, the computing device 700 does not need to include a PCI bus.

[0161] Memory 704 may include any of a variety of computer-readable media. Computer-readable media can be any available media that can be accessed by computing device 700. Computer-readable media can include volatile and non-volatile media and removable and non-removable media. By way of example and not limitation, computer-readable media can include computer storage media and communication media.

[0162] Computer storage media may include 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 704 may store computer-readable instructions (e.g., instructions representing one or more programs and / or one or more program elements), such as an operating system. Computer storage media may include, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and that can be accessed by the computing device 700. As used herein, computer storage media itself does not include signals.

[0163] Computer storage media can carry computer-readable instructions, data structures, program modules, and / or other data types contained in a modulated data signal (e.g., a carrier wave or other transport mechanism), and includes any information delivery media. The term "modulated data signal" may refer to a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, computer storage media can include wired media, such as a wired network or direct wired connection, and wireless media, such as acoustic, RF, infrared, and other wireless media. Any combination of the above should also be included within the scope of computer-readable media.

[0164] One or more CPUs 706 may be configured to execute at least some computer-readable instructions to control one or more components of the computing device 700 to perform one or more methods and / or processes described herein. One or more CPUs 706 may each include one or more cores (e.g., one, two, four, eight, twenty-eight, seventy-two, etc.) capable of processing multiple software threads simultaneously. One or more CPUs 706 may include any type of processor and may include different types of processors (e.g., processors with fewer cores for mobile devices and processors with more cores for servers) depending on the type of computing device 700 implemented. For example, depending on the type of computing device 700, 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). In addition to one or more microprocessors or supplementary coprocessors (e.g., math coprocessors), the computing device 700 may also include one or more CPUs 706.

[0165] In addition to, or instead of, the one or more CPUs 706, one or more GPUs 708 can be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 700 to perform one or more of the methods and / or processes described herein. The one or more GPUs 708 can be integrated GPUs (e.g., with the one or more CPUs 706 and / or the one or more GPUs 708 can be discrete GPUs). In embodiments, the one or more GPUs 708 can be co-processors of the one or more CPUs 706. The computing device 700 can use the one or more GPUs 708 to render graphics (e.g., 3D graphics) or perform general-purpose computing. For example, the one or more GPUs 708 can be used for general-purpose computing on GPUs (GPGPU). The one or more GPUs 708 can include hundreds or thousands of cores capable of processing hundreds or thousands of software threads simultaneously. The one or more GPUs 708 can generate pixel data for output images in response to rendering commands (e.g., rendering commands received from the one or more CPUs 706 via a host interface). The one or more GPUs 708 can include graphics memory, such as display memory, for storing pixel data or any other suitable data, such as GPGPU data. The display memory can be included as part of the memory 704. The one or more GPUs 708 can include two or more GPUs operating in parallel (e.g., over a link). The link can connect the GPUs directly (e.g., using NVLINK) or through a switch (e.g., using an NVSwitch). When combined together, each GPU 708 can generate pixel data or GPGPU data for a different portion of an output or a different output (e.g., a first GPU for a first image and a second GPU for a second image). Each GPU can include its own memory or can share memory with other GPUs.

[0166] In addition to, or in lieu of, one or more CPUs 706 and / or one or more GPUs 708, one or more logic units 720 can be configured to execute at least some computer-readable instructions to control one or more components of computing device 700 to perform one or more methods and / or processes described herein. In embodiments, one or more CPUs 706, one or more GPUs 708, and / or one or more logic units 720 can perform any combination of methods, processes, and / or portions thereof discretely or jointly. One or more of logic units 720 can be part of and / or integrated with one or more of CPUs 706 and / or GPUs 708, and / or one or more of logic units 720 can be discrete components or otherwise external to one or more CPUs 706 and / or one or more GPUs 708. One or more logic units 720 can be a co-processor of one or more CPUs 706 and / or one or more GPUs 708.

[0167] Examples of one or more logic units 720 include one or more processing cores and / or components thereof, such as a data processing unit (DPU), a tensor core (TC), a tensor processing unit (TPU), a pixel visual core (PVC), a visual processing unit (VPU), a graphics processing cluster (GPC), a texture processing cluster (TPC), a streaming multi-processor (SM), a tree traversal unit (TTU), an artificial intelligence accelerator (AIA), a deep learning accelerator (DLA), an arithmetic logic unit (ALU), an application-specific integrated circuit (ASIC), a floating-point unit (FPU), an input / output (I / O) element, a peripheral component interconnect (PCI) or peripheral component interconnect express (PCIe) element, and / or the like.

[0168] The communication interface 710 may include one or more receivers, transmitters, and / or transceivers that enable the computing device 700 to communicate with other computing devices via an electronic communication network including wired and / or wireless communications. The communication interface 710 may include components and functionality to support communication via 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., communication via 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 720 and / or the communication interface 710 may include one or more data processing units (DPUs) to transmit data received via the network and / or via the interconnect system 702 directly to one or more GPUs 708 (e.g., memory of one or more GPUs 708).

[0169] I / O ports 712 may logically couple computing device 700 to other devices, including I / O components 714, one or more presentation components 718, and / or other components, some of which may be built into (e.g., integrated into) computing device 700. Illustrative I / O components 714 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, and the like. I / O components 714 may provide a natural user interface (NUI) that processes in-air gestures, voice, or other physiological input generated by a user. In some cases, the input may be transmitted to an appropriate network element for further processing. The NUI may implement any combination of voice recognition, stylus recognition, facial recognition, biometrics, gesture recognition on and near the screen, in-air gestures, head and eye tracking, and touch recognition associated with the display of computing device 700 (described in more detail below). Computing device 700 may include a depth camera, such as a stereo camera system, an infrared camera system, an RGB camera system, touch screen technology, and combinations thereof, for gesture detection and recognition. Additionally, computing device 700 may include an accelerometer or gyroscope capable of detecting motion (e.g., as part of an inertial measurement unit (IMU)). In some examples, the output of the accelerometer or gyroscope may be used by computing device 700 to render immersive augmented or virtual reality.

[0170] The power supply 716 may include a hardwired power supply, a battery power supply, or a combination thereof. The power supply 716 may provide power to the computing device 700 to enable the components of the computing device 700 to operate.

[0171] The one or more presentation components 718 may include a display (e.g., a display, a touch screen, a television screen, a head-up display (HUD), other display types, or combinations thereof), speakers, and / or other presentation components. The one or more presentation components 718 may receive data from other components (e.g., one or more GPUs 708, one or more CPUs 706, a DPU, etc.) and output data (e.g., as images, video, sound, etc.).

[0172] Sample Data Center

[0173] Figure 8 An example data center 800 that may be used in at least one embodiment of the present disclosure is shown. The data center 800 may include a data center infrastructure layer 810, a framework layer 820, a software layer 830, and / or an application layer 840.

[0174] like Figure 8 As shown, the data center infrastructure layer 810 may include a resource coordinator 812, group computing resources 814, and node computing resources ("node CRs") 816(1)-816(N), where "N" represents any integer. In at least one embodiment, the node CRs 816(1)-816(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.), storage devices (e.g., dynamic read-only memories), and in some embodiments, 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., and one or more of the node CRs 816(1)-816(N) may correspond to servers having one or more of the above-mentioned computing resources. Furthermore, in some embodiments, the node CRs 816(1)-8161(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 816(1)-816(N) may correspond to a virtual machine (VM).

[0175] In at least one embodiment, the grouped computing resources 814 may include separate groups of node CRs 816 located in one or more racks (not shown), or multiple racks located in a data center at different geographical locations (also not shown). The separate groups of node CRs 816 within the grouped computing resources 814 may include grouped computing, network, memory, or storage resources that can be configured or allocated to support one or more workloads. In at least one embodiment, multiple node CRs 816 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. One or more racks may also include any number of power modules, cooling modules, and / or network switches in any combination.

[0176] The resource coordinator 812 may configure or otherwise control one or more node CRs 816(1)-816(N) and / or group computing resources 814. In at least one embodiment, the resource coordinator 812 may comprise a software design infrastructure (SDI) management entity for the data center 800. The resource coordinator 812 may comprise hardware, software, or some combination thereof.

[0177] In at least one embodiment, Figure 8 As shown, the framework layer 820 may include a job scheduler 832, a configuration manager 834, a resource manager 836, and / or a distributed file system 838. The framework layer 820 may include a framework that supports the software 832 of the software layer 830 and / or one or more applications 842 of the application layer 840. The software 832 or one or more applications 842 may respectively include web-based service software or applications, such as those provided by Amazon Web Services, Google Cloud, and Microsoft Azure. The framework layer 820 may be, but is not limited to, a free and open source software web application framework, such as Apache Spark™ (hereinafter referred to as "Spark"), which can utilize the distributed file system 838 for large-scale data processing (e.g., "big data"). In at least one embodiment, the job scheduler 832 may include a Spark driver to facilitate the scheduling of workloads supported by the various layers of the data center 800. The configuration manager 834 may configure the various layers, such as the software layer 830 and the framework layer 820, including Spark and the distributed file system 838, to support large-scale data processing. The resource manager 836 may be capable of managing clustered or grouped computing resources that are mapped to or allocated to support the distributed file system 838 and the job scheduler 832. In at least one embodiment, the clustered or grouped computing resources may include grouped computing resources 814 at the data center infrastructure layer 810. The resource manager 836 may coordinate with the resource coordinator 812 to manage these mapped or allocated computing resources.

[0178] In at least one embodiment, software 832 included in software layer 830 can include software used by at least portions of node C.R.s 816(1)-816(N), grouped computing resources 814, and / or distributed file system 838 of framework layer 820. One or more types of software can include, but are not limited to, internet web page search software, email virus scanning software, database software, and streaming video content software.

[0179] In at least one embodiment, applications 842 included in application layer 840 can include one or more types of applications used by at least portions of node C.R.s 816(1)-816(N), grouped computing resources 814, and / or distributed file system 838 of framework layer 820, but are not limited to any number of genomics applications, perceptual computing and machine learning applications including training or inferencing software, machine learning framework software (e.g., PyTorch, TensorFlow, Caffe, etc.), and / or other machine learning applications used in conjunction with one or more embodiments.

[0180] In at least one embodiment, any of configuration manager 834, resource manager 836, and resource orchestrator 812 can implement any number and type of self-modifying actions based on any amount and type of data acquired in any technically feasible fashion. Self-modifying actions can free data center 800’s data center operators from making potentially erroneous configuration decisions and can avoid underutilization and / or poor performance of portions of a data center.

[0181] In accordance with one or more embodiments described herein, data center 800 can include tools, services, software, or other resources for training one or more machine learning models or using one or more machine learning models to predict or infer information. For example, one or more machine learning models can be trained by computing weight parameters according to a neural network architecture using software and / or computing resources described above with respect to data center 800. In at least one embodiment, a trained or deployed machine learning model corresponding to one or more neural networks can be used to infer or predict information using resources described above with respect to data center 800 by using weight parameters computed through one or more training techniques such as, but not limited to, training techniques described herein.

[0182] In at least one embodiment, the data center 800 can use a CPU, an application-specific integrated circuit (ASIC), a GPU, an FPGA, and / or other hardware (or corresponding virtual computing resources) to perform training and / or inference using the above resources. In addition, one or more of the above software and / or hardware resources can be configured as a service to allow users to train or perform information reasoning, such as image recognition, speech recognition, or other artificial intelligence services.

[0183] Sample network environment

[0184] A network environment suitable for implementing embodiments of the present invention may include one or more client devices, servers, network attached storage (NAS), other backend devices, and / or other device types. The client devices, servers, and / or other device types (e.g., each device) may be configured to: Figure 7 The data center 800 may be implemented on one or more instances of one or more computing devices 700 - for example, each device may include similar components, features and / or functions of the computing device 700. In addition, back-end devices (servers, NAS, etc.) may be included as part of the data center 800, examples of which are referred to herein. Figure 8 Describe in more detail.

[0185] The components of the network environment can communicate with each other via one or more networks that are wired, wireless, or both. The network can include multiple networks, or networks 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 (e.g., the Internet and / or a public switched telephone network (PSTN)), and / or one or more private networks. In the case where the network includes a wireless telecommunications network, components such as base stations, communication towers, or even access points (among other components) can provide wireless connectivity.

[0186] Compatible network environments may include one or more peer-to-peer network environments, in which case the server may not be included in the network environment, and one or more client-server network environments, in which case one or more servers may be included in the network environment. In a peer-to-peer network environment, the functionality described herein with respect to the server may be implemented on any number of client devices.

[0187] In at least one embodiment, the network environment may include one or more cloud-based network environments, distributed computing environments, combinations thereof, and the like. The cloud-based network environment may include a framework layer, a job scheduler, a resource manager, and a distributed file system implemented on one or more servers, which may include one or more core network servers and / or edge servers. The framework layer may include software supporting the software layer and / or a framework for one or more applications at the application layer. The software or one or more applications may include web-based service software or applications, respectively. In an embodiment, one or more client devices may use web-based service software or applications (e.g., by accessing the service software and / or applications via one or more application programming interfaces (APIs)). The framework layer may be, but is not limited to, a free, open source software web application framework that, for example, may use a distributed file system for large-scale data processing (e.g., "big data").

[0188] A cloud-based network environment can provide cloud computing and / or cloud storage that performs any combination of the computing and / or data storage functions (or one or more portions thereof) described herein. Any of these various functions can be distributed to multiple locations from a central or core server (e.g., a server in one or more data centers), which can be distributed across a state, a region, a country, the globe, etc. If the connection to the user (e.g., a client device) is relatively close to one or more edge servers, the one or more core servers can assign at least a portion of the functions to the one or more edge servers. A cloud-based network environment can be private (e.g., limited to a single organization), public (e.g., available to many organizations), and / or a combination thereof (e.g., a hybrid cloud environment).

[0189] One or more client devices may include the Figure 7 At least some of the components, features, and functionality of one or more example computing devices 700 are described. By way of example and not limitation, a client device may be embodied as a personal computer (PC), a laptop, a mobile device, a smartphone, a tablet computer, a smartwatch, 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 camera, a surveillance device or system, a vehicle, a watercraft, an aircraft, a virtual machine, a drone, a robot, a handheld communication device, a hospital device, a gaming device or system, an entertainment system, a vehicle computer system, an embedded system controller, a remote control, an appliance, a consumer electronic device, a workstation, an edge device, any combination of the depicted devices, or any other suitable device.

[0190] The present 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). Generally, program modules include routines, programs, objects, components, data structures, etc., which refer to code that performs a specific task or implements a specific abstract data type. The present disclosure can be implemented in a variety of system configurations, including handheld devices, consumer electronics, general-purpose computers, more specialized computing devices, etc. The present invention can also be implemented in distributed computing environments, where tasks are performed by remote processing devices linked through a communications network.

[0191] As used herein, the statement "and / or" with respect to two or more elements should be interpreted as referring to only one element or a combination of elements. For example, "element A, element B, and / or element C" may include only element A, element B, element C, element A and element B, element A and element C, element B and element C, or element A, element B, and element C. In addition, "at least one of element A or element B" may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B. In addition, "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.

[0192] To satisfy statutory requirements, the subject matter of the present disclosure is described herein in detail. However, the description itself is not intended to limit the scope of the present disclosure. Rather, the inventors have contemplated that the claimed subject matter may also be embodied in other ways, including different steps or combinations of steps similar to the steps described in this document, as well as other existing or future technologies. Furthermore, although the terms "step" and / or "block" may be used herein to imply different elements of the method employed, the terms should not be interpreted as implying any particular order between the various steps disclosed herein unless the order of the steps is explicitly described.

Claims

1. A processor, comprising: One or more processing units for: detecting an object in the environment based on at least first sensor data representing an image generated at a first time using an image sensor of the ego machine; determining a first probability distribution corresponding to a position of the object in a three-dimensional 3D world space based at least in part on the detection of the object; generating a set of particles at one or more first locations in the first probability distribution, the set of particles having one or more first confidence values ​​indicating whether the object is present at the one or more first locations; generating, at a second time after the first time, a set of particles at one or more second locations based at least in part on the one or more first locations and movement of the self-machine between the first time and the second time, the set of particles at the one or more second locations including one or more second confidence values ​​indicating whether the object is present at the one or more second locations; correlating the first sensor data with second sensor data representing one or more depth detections generated using a depth sensor of the ego machine at a third time; updating the one or more second confidence values ​​for the set of particles to one or more third confidence values ​​indicating whether the object is present at the one or more second locations based at least in part on the correlation; as well as One or more particles in the set of particles having associated one or more third confidence values ​​less than a threshold confidence value are filtered out to determine a subset of the particles.

2. The processor of claim 1 , wherein the one or more processing units determine one or more second probability distributions based at least in part on at least one of: an estimated sensor calibration error between the image sensor and the depth sensor or one or more estimated errors in the prediction of the motion of the ego machine.

3. The processor of claim 1 , wherein the one or more processing units correlate the first sensor data with the second sensor data by: determining a first timestamp corresponding to the first sensor data and a second timestamp corresponding to the second sensor data; and The depth detection is aligned with the image based at least in part on the motion of the ego machine between the first time stamp and the second time stamp. 4 . The processor of claim 3 , wherein the one or more processing units further align the depth detection with the image by using at least one of intrinsic parameters or extrinsic parameters of the image sensor and the depth sensor. 5 . The processor of claim 1 , wherein the one or more processing units further utilize the subset of the particles to perform one or more operations.

6. The processor of claim 1 , wherein the one or more processing units correlate the first sensor data with the second sensor data by: projecting the one or more depth detections onto the image as one or more second probability distributions; and At least one set of the one or more depth detections associated with detection of the object in the image is determined based at least in part on the projections.

7. A processor according to claim 6, wherein the detection of the object is represented using a bounding shape, and the one or more processing units determine the at least one group of the one or more depth detections associated with the detection of the object in the image by determining which of the one or more second probability distributions are within the bounding shape.

8. The processor of claim 1, wherein the second confidence value differs from the first confidence value based at least in part on an estimated error corresponding to motion of the ego-machine.

9. The processor of claim 1 , wherein the processor is included in at least one of: control systems for autonomous or semi-autonomous machines; Perception systems for autonomous or semi-autonomous machines; a system for performing simulation operations; Systems for performing deep learning operations; Systems implemented using edge devices; Systems implemented using robots; A system comprising one or more virtual machines VM; A system implemented at least in part in a data center; or A system implemented at least in part using cloud computing resources.

10. A system comprising: Image sensor; Depth sensor; one or more processing units; as well as one or more memory units storing instructions that, when executed by the one or more processing units, cause the one or more processing units to perform operations comprising: determining one or more second-particle positions of one or more particles corresponding to an object at a first time and a motion of the ego-machine between the first and second times, the particles having one or more associated first confidences indicating whether the object is located at the one or more second-particle positions, the one or more first-particle positions being determined using first sensor data generated using the image sensor; correlating the first sensor data generated using the image sensor with second sensor data generated using the depth sensor; comparing one or more depth detections from the second sensor data to a boundary shape corresponding to the object represented by the first sensor data based at least in part on the correlation; determining a set of the one or more depth detections within the bounding shape; updating the one or more associated first confidence levels to one or more second confidence levels based at least in part on the set of one or more depth detections; and The position of the object is determined using the one or more second confidence levels. 11 . The system of claim 10 , wherein the one or more first particle positions and the one or more second particle positions are represented in a three-dimensional (3D) world space. 12 . The system of claim 10 , wherein the comparing comprises projecting the one or more depth detections from a three-dimensional (3D) world space to a two-dimensional (2D) image space and performing the comparing in the 2D image space.

13. The system of claim 12, wherein the one or more depth detections are projected as a probability distribution function.

14. The system of claim 10, wherein determining the location of the object comprises filtering out one or more of the particles having one or more associated second confidences below a threshold.

15. The system of claim 10, wherein the one or more first particle positions are determined using a probability distribution function generated by detection of the object in an image generated using the image sensor.

16. The system of claim 10, wherein the system is included in at least one of the following: control systems for autonomous or semi-autonomous machines; Perception systems for autonomous or semi-autonomous machines; a system for performing simulation operations; Systems for performing deep learning operations; Systems implemented using edge devices; Systems implemented using robots; A system comprising one or more virtual machines VM; A system implemented at least in part in a data center; or A system implemented at least in part using cloud computing resources.

17. A method comprising: detecting an object in the environment based on at least first sensor data representing an image generated at a first time using an image sensor of the ego machine; determining a first probability distribution corresponding to a position of the object in a three-dimensional 3D world space based at least in part on detecting the object; generating a set of particles at one or more first locations in the first probability distribution, the set of particles having one or more first confidence values ​​indicating whether the object is present at the one or more first locations; generating, at a second time after the first time, a set of particles at one or more second locations based at least in part on the one or more first locations and movement of the self-machine between the first time and the second time, the set of particles at the one or more second locations including one or more second confidence values ​​indicating whether the object is present at the one or more second locations; correlating the first sensor data with second sensor data representing one or more depth detections generated using a depth sensor of the ego machine at a third time; updating the one or more second confidence values ​​for the set of particles to one or more third confidence values ​​indicating whether the object is present at the one or more second locations based at least in part on the correlation; as well as One or more particles in the set of particles having one or more associated third confidence values ​​less than a threshold confidence value are filtered out to determine a subset of the particles.

18. The method of claim 17, wherein one or more second probability distributions are determined based at least in part on at least one of an estimated sensor calibration error between the image sensor and the depth sensor or an estimated error in a prediction of the ego machine's motion.

19. The method of claim 17, wherein correlating the first sensor data with the second sensor data comprises: determining a first timestamp corresponding to the first sensor data and a second timestamp corresponding to the second sensor data; as well as The depth detection is aligned with the image based at least in part on the motion of the ego machine between the first time stamp and the second time stamp.

20. The method of claim 17, wherein correlating the first sensor data with the second sensor data comprises: projecting the one or more depth detections onto the image as one or more second probability distributions; as well as At least one set of the one or more depth detections associated with detection of the object in the image is determined based at least in part on the projections.

21. A processor comprising: processing circuitry for performing one or more control operations corresponding to the machine based at least on a location of a hazard corresponding to a set of particles at a first time, The position is determined based on at least one or more confidence values ​​indicating whether an object is present at one or more particle positions associated with the group of particles at a second time, the one or more confidence values ​​being determined based on at least one or more depth detections associated with the one or more particle positions, using the one or more depth detections and the one or more particle positions, The one or more particle positions correspond to at least one image sensor associated with the machine, and the one or more depth detections correspond to at least one depth sensor associated with the machine.

Images