Determining positioning accuracy in autonomous and semi-autonomous systems and applications

Abstract

Determining positioning accuracy in autonomous and semi-autonomous systems and applications is disclosed. In various examples, accuracy determination for positioning in autonomous and semi-autonomous systems and applications is described herein. Systems and methods for determining one or more errors associated with vehicle positioning using various types of sensor data generated with a vehicle are disclosed. For example, a first component of a vehicle may determine an estimated attitude of the vehicle using a map and first sensor data. A second component of the vehicle may then determine one or more errors associated with the estimated pose based on actual motion of the vehicle within the environment (as determined using the second sensor data) and comparing features represented by the first sensor data to features represented by the map. In some examples, the second component may further determine information associated with the error, such as one or more uncertainties associated with the error.

Description

Background Art

[0001] Mapping and localization are common processes for autonomous and semi-autonomous driving functions. High-definition (HD) maps, sensor perception, or a combination thereof are typically used to locate the vehicle relative to the HD map in order to make planning and control decisions. For example, the vehicle may generate sensor data using one or more sensors, such as one or more image sensors, one or more laser radar (LiDAR) sensors, one or more radar (RADAR) sensors, etc. The vehicle may then compare the features represented by the sensor data with the features represented by the HD map. Based on the comparison, the vehicle may determine a posture associated with the vehicle within the environment, such as the position and / or orientation of the vehicle within the environment. However, even if this positioning process can provide accurate results, there may still be errors associated with the determined posture. In some examples, this error may depend on one or more factors, such as noise, the number and / or type of features represented by the sensor data, whether the sensor data is obscured, whether the HD map is completed in the area of ​​the environment in which the vehicle is navigating, and / or other factors.

[0002] Therefore, some technologies have been developed to estimate one or more errors between the positioning attitude of a vehicle and the actual attitude of the vehicle. For example, during a time window, a system can use the motion of the vehicle during the time window to determine the previous attitude of the vehicle when the time window begins and then determine the final attitude of the vehicle when the time window ends. The system can then use the final attitude and the positioning attitude to determine one or more errors associated with positioning when the time window ends. For example, the system can determine one or more errors based on one or more differences between the final attitude and the positioning attitude. Although this technology does provide an estimation of one or more errors, one or more errors can change based on the size of the time window, which can make one or more errors less relevant. In addition, this technology does not allow the use of error boundaries associated with one or more errors. Summary of the invention

[0003] Embodiments of the present disclosure relate to accuracy determination of positioning for autonomous and semi-autonomous systems and applications. Systems and methods for determining one or more errors associated with vehicle positioning using various types of sensor data generated using a vehicle are disclosed. For example, a first component of a vehicle may use a map and first sensor data (such as image data, LiDAR data, and / or RADAR data) to determine an estimated posture of the vehicle. A second component of the vehicle may then determine the error associated with the estimated posture based on the actual movement of the vehicle within the environment (as determined using the second sensor data) and comparing the features represented by the first sensor data with the features represented by the map. In some examples, the second component may further determine information associated with (one or more) errors, such as one or more uncertainties associated with (one or more) errors. In addition, in some examples, (one or more) systems may perform one or more processes based on one or more errors, such as determining whether to use the estimated posture when navigating.

[0004] Compared to conventional systems (such as those described above), in some embodiments, the current system determines one or more errors associated with vehicle positioning by fusing different types of sensor data (such as motion data representing the actual movement of the vehicle and image data, LiDAR data, RADAR data, etc.). For example, the current system can propagate errors determined by using motion data and errors determined using image data, LiDAR data, RADAR data, etc., in order to determine one or more final errors associated with vehicle positioning, which may be more accurate than using motion alone. In addition, compared to conventional systems, in some embodiments, the current system does not need to use factors that can change the error, such as a time window. Instead, the current system can continuously determine and / or update one or more errors in vehicle positioning when new data (e.g., sensor data) is received from the vehicle's sensors. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] The present system and method for determining accuracy of positioning in autonomous and semi-autonomous systems and applications is described in detail below with reference to the accompanying drawings, in which:

[0006] Figure 1 An example data flow diagram illustrating a process for determining an error associated with vehicle positioning according to some embodiments of the present disclosure;

[0007] Figure 2A A data flow diagram showing a process for performing positioning according to some embodiments of the present disclosure is shown;

[0008] Figure 2B The relationship between the vehicle origin, the road segment origin and the global origin used for positioning according to some embodiments of the present disclosure is shown;

[0009] Figure 3A shows a sampled cost space corresponding to a frame generated during a positioning process according to some embodiments of the present disclosure;

[0010] Figure 3B shows a total cost space corresponding to a sampled cost space of a plurality of frames generated during a positioning process according to some embodiments of the present disclosure;

[0011] Figure 3C shows a filtered total cost space generated during the positioning process according to some embodiments of the present disclosure;

[0012] FIG. 4A to FIG. 4C shows an example visualization for sampling poses of a sampling cost space based on camera positioning according to some embodiments of the present disclosure;

[0013] FIG. 5A to FIG. 5B shows an example visualization for sampling poses of a sample cost space for RADAR-based positioning according to some embodiments of the present disclosure;

[0014] FIG. 6A to FIG. 6D shows an example visualization for sampling poses of a sample cost space for LiDAR-based positioning according to some embodiments of the present disclosure;

[0015] Figure 7 An example of fusing multiple positioning results to generate a final positioning result according to some embodiments of the present invention is shown;

[0016] Figure 8 illustrates an example of one or more errors between an actual pose associated with a vehicle and a positioning pose associated with the vehicle according to some embodiments of the present invention;

[0017] Fig. 9 An example of using motion associated with a vehicle to determine a true pose associated with a vehicle is shown in accordance with some embodiments of the present disclosure;

[0018] Fig.10 An example of determining one or more errors associated with the motion of a vehicle according to some embodiments of the present invention is shown;

[0019] Fig.11 An example of using both the relative motion of the vehicle and positioning corrections to determine one or more errors associated with the motion of the vehicle is shown in accordance with some embodiments of the present disclosure;

[0020] FIG. 12A to FIG. 12B An example of determining an error associated with positioning of a vehicle according to some embodiments of the present disclosure is shown;

[0021] Fig.13 A flow chart illustrating a method for determining one or more errors associated with positioning according to some embodiments of the present disclosure;

[0022] Fig.14 A flow chart illustrating a method for determining an operating mode based at least on one or more errors associated with positioning according to some embodiments of the present disclosure;

[0023] Fig.15A is an illustration of an example autonomous vehicle according to some embodiments of the present disclosure;

[0024] Fig. 15B According to some embodiments of the present disclosure Fig.15A Examples of autonomous vehicle camera positions and fields of view;

[0025] Fig. 15C According to some embodiments of the present disclosure Fig.15A a block diagram of an example system architecture for an example autonomous vehicle;

[0026] Fig.15D According to some embodiments of the present disclosure, a method for performing a multi-threaded communication between one or more cloud-based servers and Fig.15A System diagram of an example of communication between autonomous vehicles;

[0027] Fig.16 is a block diagram of an example computing device suitable for implementing some embodiments of the present disclosure; and

[0028] Fig.17 is a block diagram of an example data center suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0029] Systems and methods are disclosed that relate to accuracy determination of positioning for autonomous and semi-autonomous systems and applications. Although the present disclosure may be described with respect to an example autonomous or semi-autonomous vehicle 1500 (or, referred to herein as “vehicle 1500,” “ego vehicle 1500,” “machine 1500,” or “ego machine 1500”), the example autonomous or semi-autonomous vehicle may be described with respect to an example autonomous or semi-autonomous vehicle 1500. FIG. 15A to FIG. 15DFor description, this is not intended to be limiting. For example, the systems and methods described herein may be used by, but are not limited to, non-autonomous vehicles or machines, semi-autonomous vehicles or machines (e.g., in one or more adaptive driver assistance systems (ADAS)), autonomous vehicles or machines, driven and undriven robots or robotic platforms, warehouse vehicles, off-road vehicles, vehicles coupled to one or more trailers, flying boats, boats, shuttles, emergency response vehicles, motorcycles, electric or motorized bicycles, airplanes, construction vehicles, underwater vehicles, drones, and / or other vehicle types. In addition, although the present disclosure may be described with respect to machine positioning, this is not intended to be limiting, and the systems and methods described herein may be used for augmented reality, virtual reality, mixed reality, robotics, security and surveillance, autonomous or semi-autonomous machine applications, and / or any other technology space where machine or other object positioning may be used.

[0030] For example, one or more systems (e.g., a first component) may use one or more techniques to locate a vehicle (and / or other type of machine) within an environment. In some examples, and as described in more detail herein, one or more systems locate a vehicle by determining an initial pose of the vehicle using sensor data (in some examples, referred to as "first sensor data") generated using one or more position sensors (such as a global positioning system (GPS)) of the vehicle. Then, one or more systems refine and / or update the initial pose of the vehicle using sensor data (in some examples, referred to as "second sensor data") generated using one or more other sensors (such as one or more image sensors (e.g., one or more cameras), one or more LiDAR sensors, one or more RADAR sensors, etc.) of the vehicle. For example, to refine the initial pose, one or more systems may compare one or more features represented by the second sensor data with one or more features represented by map data, wherein the map data represents a map of the environment in which the vehicle is located. Based at least on the comparison, one or more systems may update the initial pose of the vehicle to an estimated pose (e.g., a positioning pose) of the vehicle within the environment. As described herein, a pose may represent a vehicle's position (e.g., an x-coordinate position, a y-coordinate position, and / or a z-coordinate position), a vehicle's orientation (e.g., yaw, pitch, and / or roll), and / or any other position, pose, or orientation information.

[0031] In some examples, the positioning of the vehicle (e.g., the estimated position and / or attitude) may include one or more errors based at least on one or more factors, such as noise associated with the positioning process, the second sensor data not representing a sufficient number of features (e.g., the vehicle is navigating through a tunnel), the second sensor data being blocked (e.g., based on weather conditions), an incomplete map of the environment the vehicle is navigating, etc. For example, as described herein, the one or more errors may include a first error between the estimated attitude and the actual attitude of the vehicle in a first coordinate direction (e.g., the x-coordinate direction), a second error between the estimated attitude and the actual attitude of the vehicle in a second coordinate direction (e.g., the y-coordinate direction), a third error between the estimated attitude and the actual attitude of the vehicle in a third coordinate direction (e.g., the z-coordinate direction), a fourth error of a yaw difference between the estimated attitude and the actual attitude of the vehicle, a fifth error of a roll difference between the estimated attitude and the actual attitude of the vehicle, a sixth error of a pitch difference between the estimated attitude and the actual attitude of the vehicle, etc. As such, one or more systems (e.g., the second component) may determine the one or more errors associated with positioning using one or more techniques.

[0032] For example, one or more systems may use a previous posture associated with the vehicle and sensor data generated using one or more motion sensors (e.g., one or more inertial measurement unit (IMU) sensors, etc.) (in some examples, referred to as "third sensor data") to determine the true posture associated with the vehicle. In some examples, the posture may be considered to be a "true" posture based on the assumption that the motion data is accurate and / or substantially accurate for at least a short period of time and / or a short travel distance associated with the vehicle. As such, in some examples, the true posture may include and / or correspond to the actual posture of the vehicle. One or more systems may then determine one or more differences between the true posture and the estimated posture, such as one or more position differences (e.g., differences in x-coordinate direction, y-coordinate direction, and / or z-coordinate position) and / or one or more orientation differences (e.g., differences in yaw, pitch, and / or roll). In some examples, one or more systems may use one or more differences to determine one or more errors (in some examples, referred to as "(one or more) first errors").

[0033] The one or more systems may also compare one or more features represented by the second sensor data with one or more features represented by the map data. Based at least on the comparison, the one or more systems may determine one or more correspondences between the (one or more) features represented by the second sensor data and the (one or more) features represented by the map data. In some examples, the one or more systems perform the comparison and / or determination using one or more techniques similar to those used when performing positioning. For example, in some examples, the correspondence may indicate whether the feature represented by the second sensor data matches the feature represented by the map data, whether the location of the feature represented by the second sensor data matches the location of the same feature represented by the map data, whether the cost associated with the feature represented by the second sensor data matches the feature represented by the map data, the cost associated with the gesture based on the feature matching, and / or any other relationship based at least on the feature matching. However, in some examples, the one or more systems may determine one or more errors (in some examples, referred to as "(one or more) second errors") using the one or more correspondences.

[0034] Then, one or more systems may determine one or more final errors associated with the positioning based at least on one or more differences associated with the vehicle motion (e.g., (one or more) first errors) and one or more correspondences associated with the vehicle positioning (e.g., (one or more) second errors). In some examples, one or more systems may determine the one or more final errors using one or more algorithms, such as one or more algorithms associated with a Kalman filter and / or other types of estimates. In some examples, one or more systems may provide greater weight to the (one or more) differences associated with the vehicle and / or the (one or more) correspondences associated with the comparison when determining the (one or more) final errors. Additionally, in some examples, one or more systems may determine one or more uncertainties associated with the (one or more) final errors, for example by using one or more algorithms.

[0035] When the vehicle navigates around the environment, one or more systems can continue to perform these processes. For example, one or more systems can continue to perform these processes after a given time period (such as but not limited to every 10 milliseconds, every 30 milliseconds, every 50 milliseconds, every second and / or using any other time period) to determine (one or more) final errors. In addition, when performing these processes for new iterations, one or more systems can use one or more final errors from previous iterations, such as by continuing to update one or more final errors when the vehicle is navigating. In addition, in some examples, one or more systems can use one or more final errors to perform one or more processes when navigating. For example, if one or more final errors are greater than one or more error thresholds, one or more systems can cause the vehicle to operate in a first mode (e.g., an initialization mode), in which the estimated posture may not be trusted and / or used by one or more other systems of the vehicle. In addition, if the (one or more) final errors are less than or equal to the (one or more) error thresholds, one or more systems can cause the vehicle to operate in a second mode in which the estimated pose can be trusted and / or used by (one or more) other systems of the vehicle.

[0036] The systems and methods described herein may be used by, but are not limited to, non-autonomous vehicles or machines, semi-autonomous vehicles or machines (e.g., in one or more adaptive driver assistance systems (ADAS)), autonomous vehicles or machines, manned and unmanned robots or robotic platforms, warehouse vehicles, off-road vehicles, vehicles coupled to one or more trailers, flying boats, boats, shuttles, emergency response vehicles, motorcycles, electric or motorized bicycles, airplanes, construction vehicles, underwater vehicles, drones, and / or other vehicle types. Further, the systems and methods described herein may be used for a variety of purposes, including, by way of example and not limitation, for machine control, machine motion, machine travel, synthetic data generation, model training, perception, augmented reality, virtual reality, mixed reality, robotics, security and surveillance, simulation and digital twins, autonomous or semi-autonomous machine applications, generative AI applications, deep learning, environmental simulation, object or participant simulation and / or digital twins, data center processing, conversational AI, light transport simulation (e.g., ray tracing, path tracing, etc.), collaborative content creation for 3D assets, cloud computing, and / or any other suitable application.

[0037] The disclosed embodiments may be included in a variety of different systems, such as automotive systems (e.g., control systems for autonomous or semi-autonomous machines, perception systems for autonomous or semi-autonomous machines), systems implemented using robots, aerial systems, medical systems, boating systems, smart area monitoring systems, systems for performing deep learning operations, systems for performing simulation operations, systems for performing digital twin operations, systems implemented using edge devices, systems implementing one or more language models - such as one or more large language models (LLMs), systems including one or more virtual machines (VMs), systems for performing synthetic data generation operations, systems implemented at least in part in a data center, systems for performing conversational AI operations, systems for performing one or more generative AI operations, systems for performing light transport simulations, systems for performing collaborative content creation of 3D assets, systems implemented at least in part using cloud computing resources, and / or other types of systems.

[0038] refer to Figure 1 , Figure 1 An example data flow diagram of a process 100 for determining an error associated with the positioning of a vehicle in accordance with some embodiments of the present disclosure is shown. It should 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, orders, 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. Further, many of the elements described herein are functional entities that may be implemented as discrete or distributed components or in combination with other components, and in any suitable combination and location. The various functions described herein as being performed by an entity may be performed by hardware, firmware, and / or software. For example, the various functions may be implemented by a processor executing instructions stored in a memory. In some embodiments, the systems, methods, and processes described herein may use FIG. 15A to FIG. 15D Example of autonomous vehicle 1500, Fig.16 The example computing device 1600 of FIG. 1600 , and / or Fig.17 The example data center 1700 may be implemented using similar components, features, and / or functions to those of the example data center 1700.

[0039] The process 100 may include a positioning component 102 that uses the sensor data 104 and the map data 106 to locate the vehicle within the environment. In some examples, the positioning component 102 locates the vehicle by determining an initial pose of the vehicle using at least a portion of the sensor data 104 (such as location data generated by GPS). The positioning component 102 may then refine and / or update the initial pose of the vehicle using at least a second portion of the sensor data 104, such as image data generated using one or more image sensors (e.g., one or more cameras), LiDAR data generated using one or more LiDAR sensors, RADAR data generated using one or more RADAR sensors, and the like. For example, to refine the initial pose, the positioning component 102 may compare one or more features represented by the sensor data 104 with one or more features represented by the map data 106, wherein the map data 106 represents a map of the environment in which the vehicle is located. Based at least on the comparison, the positioning component 102 may update the initial pose of the vehicle to an estimated pose (e.g., a positioning pose) of the vehicle within the environment. As described herein, a pose may represent a vehicle's position (e.g., an x-coordinate position, a y-coordinate position, and / or a z-coordinate position), a vehicle's orientation (e.g., yaw, pitch, and / or roll), and / or any other positional information.

[0040] To go into more detail, Figure 2A A data flow diagram of a process 200 for performing positioning according to some embodiments of the present disclosure is depicted. In some examples, the process 200 can be performed using a vehicle (e.g., the positioning component 102). In some examples, one or more parallel processing units can be used to perform one or more of the processes described with respect to the process 200 in parallel. For example, positioning using different map layers can be performed in parallel with one or more other map layers. Within a single map layer, cost space sampling 202, cost space aggregation 204, and / or filtering 206 can be performed in parallel. For example, different poses can be sampled in parallel during cost space sampling 202 to more efficiently generate a cost space for the current frame or time step. In addition, because the map data 208 corresponding to the map layer (which can represent and / or include the map data 106) can be stored as a texture on the GPU, a texture lookup can be performed to quickly determine the cost value of the cost space, resulting in a reduction in the running time of each cost space analysis.

[0041] Figure 2B The goal of the localization process 200 shown in FIG. 200 may be to locate the origin 210 of the vehicle relative to the local origin 212 of the road segment 214 of the fused HD map represented by the map data 208. For example, the localization process 200 may be used to determine the origin 210 of the vehicle at a particular time step or frame corresponding to the road segment 214 of the fused HD map represented by the map data 208. FIG. 3A to FIG. 3C(One or more) ellipsoids of the fused positioning 216 are described in more detail. The origin 210 of the vehicle may correspond to a reference point or origin of the vehicle, such as the center of the rear axle of the vehicle. The vehicle may be positioned relative to the origin 212 of the road segment 214, and the road segment 214 may have a corresponding position 218 in the global coordinate system. In this way, once the vehicle is positioned to the road segment 214 of the HD map, the vehicle may also be globally positioned. In some examples, an ellipsoid may be determined for each individual sensor modality (e.g., LiDAR positioning, camera positioning, RADAR positioning, etc.), and the output of each positioning technology may be fused via positioning fusion 220. A final origin location - e.g., the origin 210 of the vehicle - may be generated and used as the positioning result for the vehicle at the current frame or time step.

[0042] At the beginning of the drive, the current road segment 214 of the vehicle can be determined. In some examples, the current road segment 214 may be known from the last drive - for example, when the vehicle is turned off, the last known road segment to which the vehicle is located may be stored. In other examples, the current road segment 214 may be determined. In order to determine the current road segment 214, GNSS data may be used to globally locate the vehicle, and then determine the road segment 214 corresponding to the global positioning result. In the case where the result returns two or more road segments 214, the road segment 214 with the origin 212 closest to the origin of the vehicle may be determined as the current road segment 214. Once the current road segment 214 is determined, the road segment 214 may be determined as a seed road segment for a breadth-first search. A breadth-first search may be performed to generate a local layout of road segments 214 adjacent to the current road segment 214 at a first level, then generate a local layout of road segments 214 of a second level adjacent to the road segment 214 from the first level, and so on. It should be understood that the road segments 214 adjacent to the current road segment 214 may be useful for the positioning process 200 because when the vehicle moves from one road segment 214 to another road segment 214, the relative transformation between the road segments 214 can be used to update the sampled cost spaces generated for the previous road segments 214, which are used in the total cost space used for positioning. Once the vehicle moves from the seed road segment to the adjacent road segment, another breadth-first search can be performed on the new road segment to generate an updated local layout, and the process can be repeated as the vehicle traverses the map from road segment to road segment. In addition, as described herein, when the vehicle moves from one road segment 214 to another road segment 214, the previously calculated cost space (e.g., a certain number of previous cost spaces in a buffer, such as 50, 100, etc.) can be updated to reflect the same cost space, but relative to the origin 212 of the new road segment 214. Therefore, the calculated cost space can be carried by the road segment to generate a total cost space via cost space aggregation 204.

[0043] The localization process 200 may use sensor data 222 (which may represent and / or include sensor data 104) - e.g., real-time sensor data 222 generated by the vehicle - map data 208, and / or output 228 to localize the vehicle at each time step or frame. For example, the sensor data 222, output 228, and map data 208 may be used to perform cost space sampling 202. The cost space sampling 202 may be different for different sensor modalities corresponding to different map layers. For example, the cost space sampling 202 may be performed separately for a LiDAR map layer (e.g., a LiDAR point cloud layer, a LiDAR map image layer, and / or a LiDAR voxel map layer), a RADAR map layer (e.g., a RADAR point cloud layer and / or a RADAR map image layer), and / or for a base layer (e.g., for a landmark or camera-based map layer). Within each sensor modality, one or more different techniques may be used to perform cost sampling, and the costs of the different techniques may be weighted to produce a final cost for the sampled pose. This process may be repeated for each pose in the cost space to generate the final cost space for the frame during positioning. For example, the LiDAR intensity cost, LiDAR elevation cost, and LiDAR (slice) point cloud cost (e.g., using a distance function) may be calculated, then averaged or otherwise weighted, and used for the final cost of the pose or point on the cost space. Similarly, for camera- or landmark-based cost space sampling, the semantic cost may be calculated, and the geometric cost (e.g., using a distance function) may be calculated, then averaged or otherwise weighted, and used for the final cost of the pose or point on the cost space. As another example, the RADAR point cloud cost (e.g., using a distance function) may be calculated and used for the final cost of the pose or point on the cost space. In this manner, cost space sampling 202 may be performed to sample the cost of each different pose within the cost space. The cost space may correspond to an area in the map that may include only a portion of the current segment 214, the entire current segment 214, the current segment 214 and one or more adjacent segments 214, and / or some other area of ​​the entire fused HD map. As such, the size of the cost space may be a programmable parameter of the system.

[0044] For any individual sensor modality, the result of cost space sampling 202 can be a cost space representing a geometric match or likelihood that the vehicle can be positioned relative to each particular pose. For example, a point in the cost space can have a corresponding relative position relative to the current road segment 214, and the cost space can indicate the likelihood or likelihood that the vehicle is currently in each particular pose (e.g., the (x, y, z) position relative to the origin of the road segment 214 and the axis angles around each of the x, y, and z axes).

[0045] refer to Figure 3A, cost space 302 may represent a cost space for a sensor modality—for example, based on FIG. 4A to FIG. 4C The camera-based cost space generated. For example, the cost space 302 can represent the likelihood that the vehicle is currently in each of a plurality of poses (e.g., represented by a point in the cost space 302) at the current frame. Figure 3A 2D in the figure, but the cost space 302 can correspond to a 3D cost space (e.g., with (x, y, z) positions and / or axis angles for each of the x, y, and z axes). As such, for each of the plurality of poses, the sensor data 222 (e.g., before or after preprocessing) and / or the output 228 (e.g., detection of landmark locations in the 2D image space and / or 3D image space) can be compared to the map data 208. In the event that the pose does not match the map data 208 well, the cost may be high, and the point in the cost space corresponding to the pose may be represented as such—e.g., in red, or with respect to Figure 3A , represented by the non-dashed or white portion. In the case where the pose matches the map data 208 well, the cost can be low, and the point in the cost space corresponding to the pose can be represented as such - for example, in green, or relative to Figure 3A , represented by dashed dots. For example, refer to Figure 3A , in the cost space 302 corresponds to Figure 4A In the case of visualization 402, the dashed portion 304 may correspond to low costs for poses along the diagonal where the marker 404 may match well with the prediction or output 228 of the DNN 224. For example, at the pose in the lower left portion of the dashed portion of the cost space, the prediction of the symbol may align well with the marker 404 from the map data 208, and similarly, at the upper right portion of the dashed portion, the prediction of the symbol from the corresponding pose may also align well with the marker 404. In this manner, these points may be represented at low cost. However, due to noise and a large number of low-cost poses, a single cost space 302 may not be accurate for positioning—e.g., a vehicle may not be located at every pose represented by the dashed portion 304. As such, a total cost space 310 may be generated via cost space aggregation 204, as described herein.

[0046] Cost space sampling 202 may be performed separately for different sensor modalities, as described herein. FIG. 4A to FIG. 4C, a camera or landmark-based cost space (e.g., corresponding to a base layer of a fused HD map) may be generated using geometric cost and / or semantic cost analysis at each pose in the cost space. For example, at a given time step or frame, output 228 may be calculated relative to an image (e.g., the image represented in visualization 401) - e.g., landmark locations of lane dividers, road boundaries, signs, poles, etc. 3D landmark information from map data 208 may be projected into 2D image space to correspond to the currently predicted locations of 3D landmarks in 2D image space relative to the current pose sampled by the vehicle in cost space. For example, landmark 404 may correspond to a 2D projection from map data, while landmark 406 may correspond to a current prediction or output 228 from one or more DNNs 224. Similarly, lane dividers 1014 may correspond to a 2D projection from map data 208, and lane dividers 408 may correspond to a current prediction or output 228 from one or more DNNs 224. To calculate the cost of the current pose (e.g., represented by pose indicator 410), the current output 228 from DNN 224 can be converted into a distance function (e.g., where the prediction is divided into a plurality of points, and each point has a zero cost at its center, and the cost increases moving outward from the center until a maximum cost is reached, as represented by the white area of ​​visualization 412), corresponding to the predicted geometry as represented in visualization 412, and the current output 228 can be separately converted into a semantic label for the prediction as represented in visualization 414. In addition, a 2D projection of the 3D landmark can be projected into image space, and each point from the 2D projection can be compared to the portion of the distance function representation on which the projected point lands to determine the associated cost. As such, the dashed line portion can correspond to the distance function representation of the current prediction of one or more DNNs 224, and the dark solid line or point can represent the 2D projection from map data 208. A cost can be calculated for each point of the 2D projection, and the relative cost from each point can be used to determine an average cost. For example, the cost at point 416 may be high or maximum, and the cost at point 418 may be low—for example, because the cost at point 418 is aligned with the center of the distance function representation of the lane divider. This cost may correspond to the geometric cost of the current pose for the current frame. Similarly, the semantic label corresponding to the 2D projected point from the map data 208 may be compared to the projected semantic information, such as Figure 4CAs shown. In this way, in the case where the points do not match semantically, the cost can be set to a maximum, and in the case where the points match, the cost can be set to a minimum. These values ​​can be averaged or otherwise weighted to determine the final semantic cost. The final semantic cost and the final geometric cost can be weighted to determine the final total cost for updating the cost space (e.g., cost space 302). For example, for each point, the semantic cost may have a vote that must be lower than the corresponding geometric cost. In this way, in the case where the semantic information does not match, the cost of the particular point can be set to a maximum. In the case where the semantic information matches, for the semantic cost, the cost can be set to a minimum value or 0, and the final cost of the point can represent the geometric cost. Finally, the point in the cost space corresponding to the current posture can be updated to reflect the final cost of all 2D projected points.

[0047] As another example, regarding FIG. 5A to FIG. 5B , a RADAR-based cost space (e.g., a RADAR layer corresponding to a fused HD map) can be generated using a distance function corresponding to the map data 208 (e.g., a top-down projection of the RADAR point cloud, where each point is converted to a distance function representation). For example, at a given time step or frame, the map data 208 corresponding to the RADAR point cloud (as represented in visualization 502) can be converted to a distance function (as represented in visualization 504), such as where each RADAR point can have a zero cost at its center, where the cost increases to a maximum cost as the distance from the center increases. For example, with respect to visualization 504, the white portion of visualization 502 can correspond to the maximum cost. In order to calculate the cost of the current pose (e.g., represented by pose indicator 506), the RADAR data from the sensor data 222 can be converted to a RADAR point cloud and compared to the distance function representation of the RADAR point cloud (e.g., as represented in visualization 504). The hollow circles in visualizations 502 and 504 can correspond to the current RADAR point cloud prediction of the vehicle. As such, for each current RADAR point, a cost may be determined by comparing each current RADAR point to the distance function RADAR value to which the current RADAR point corresponds or lands. As such, current RADAR point 508 may have a maximum cost, while current RADAR point 510 may have a low cost—for example, because point 510 lands close to the center of the points from the RADAR point cloud in the map data 208. Ultimately, an average or other weighting of each of the costs from the current RADAR points may be calculated, and the final cost value may be used to update the cost space for the currently sampled pose.

[0048] As another example, relative to FIG. 6A to FIG. 6D, a LiDAR-based cost space (e.g., corresponding to a LiDAR layer of a fused HD map) may be generated using distance functions on (sliced) LiDAR point clouds, LiDAR intensity maps, and / or LiDAR elevation maps. For example, for a given pose as indicated by pose indicator 602, current or real-time LiDAR data (e.g., corresponding to sensor data 222) may be generated and converted to values ​​for comparison with an intensity map generated from map data 208 (e.g., as shown in visualization 604), converted to values ​​for comparison with an elevation map generated from map data 208 (e.g., as shown in visualization 606), and a LiDAR point cloud of map data 208 (e.g., as shown in visualization 608) may be converted to a distance function representation of values ​​for comparison with a current LiDAR point cloud corresponding to sensor data 222 (e.g., as shown in visualization 610). In some examples, the LiDAR point cloud may correspond to a slice of the LiDAR point cloud, and one or more individual slices may be converted to a distance function representation and used to compute the cost. The costs from the elevation comparison, intensity comparison, and distance function comparison may be averaged or otherwise weighted to determine a final cost corresponding to the current pose on the LiDAR-based cost map.

[0049] For example, about Fig. 6A , the LiDAR layer of the fused HD map represented by the map data 208 may include a LiDAR intensity (or reflectivity) image (e.g., a top-down projection of the intensity values ​​from the fused LiDAR data). For example, a painted surface (such as a lane marking) may have a higher reflectivity, and this reflected intensity may be captured and used to compare the map data 208 with the current LiDAR sensor data. The current LiDAR sensor data from the vehicle may be converted into a LiDAR intensity representation 612 and compared with the LiDAR intensity image from the map data 208 at the current pose. For points of the current LiDAR intensity representation 612 that have intensity values ​​similar to or matching those of points from the map data 208, the cost may be lower, and in the case where the intensity values ​​do not match, the cost may be higher. For example, a zero difference in the intensity of a point may correspond to a zero cost, a threshold difference and above may correspond to a maximum cost, and between a zero difference and a threshold difference, the cost may increase from a zero cost to a maximum cost. The cost of each point of the current LiDAR intensity representation 612 may be averaged or otherwise weighted with each other point to determine the cost of the LiDAR intensity comparison.

[0050] As another example, relative to Figure 6B, the LiDAR layer of the fused HD map represented by the map data 208 may include a LiDAR elevation image (e.g., a top-down projection of elevation values, resulting in a top-down depth map). The current LiDAR sensor data from the vehicle may be converted into a LiDAR elevation representation 614 and compared to the LiDAR elevation image generated from the map data 208 at the current pose. For points of the current LiDAR elevation representation 614 that have elevation values ​​that are similar or matching to points from the map data 208, the cost may be lower, and where the elevation values ​​do not match, the cost may be higher. For example, a zero difference in the elevation of a point may correspond to a zero cost, a threshold difference and above may correspond to a maximum cost, and between zero and the threshold difference, the cost may increase from zero cost to a maximum cost. The cost of each point of the current LiDAR elevation representation 614 may be averaged or otherwise weighted with each other point to determine the cost of the LiDAR elevation comparison.

[0051] In some examples, because the elevation values ​​from the map data 208 can be determined relative to the origin 212 of the current road segment 214, and the elevation values ​​from the current LiDAR elevation representation 614 can correspond to the origin 210 or reference point of the vehicle, a transformation can be performed to compare the values ​​from the LiDAR elevation representation 614 to the LiDAR elevation image from the map data 208. For example, where a point from the LiDAR elevation representation 614 has an elevation value of 1.0 meters (e.g., 1.0 meters high from the origin of the vehicle), a point in the map data 208 corresponding to the point from the representation 614 has a value of 1.5 meters, and the elevation difference between the origin 210 of the vehicle and the origin 212 of the road segment 214 is 0.5 meters (e.g., the origin 212 of the road segment is 0.5 meters higher than the origin 210 of the vehicle), the actual difference between the point from the map data 208 and the representation 614 may be 0.0 meters (e.g., 1.5 meters - 0.5 meters = 1 meter as the final value of the point from the map data 208 relative to the origin 210 of the vehicle). Depending on the example, the transformation between the values ​​of the map data 208 or the representation 614 may correspond to a transformation from the road segment origin 212 to the vehicle origin 210, from the vehicle origin 210 to the road segment origin 212, or a combination thereof.

[0052] As another example, regarding FIG. 6C to FIG. 6D, the LiDAR layer of the fused HD map represented by map data 208 may include sliced ​​LiDAR point clouds (e.g., corresponding to a ground plane slice, a giraffe plane slice, another defined slice such as a 1 meter thick slice extending 2 meters to 3 meters from the ground plane, etc.). In some examples, the point cloud may not be sliced, and the entire point cloud may be represented instead. The sliced ​​LiDAR point cloud (e.g., as shown in visualization 608) may be converted into its distance function representation (e.g., as shown in visualization 610). For example, each point from the LiDAR point cloud may be converted so that the center of the point has a zero cost, and the cost increases the farther away from the center of the point up to a certain maximum cost (e.g., represented by the white area of ​​visualization 610). In order to calculate the cost of the current pose (e.g., represented by the pose indicator), the LiDAR data from sensor data 222 may be converted into a LiDAR point cloud (or its corresponding slice) and compared with the distance function representation of the LiDAR point cloud (e.g., as represented in visualization 610). The hollow circles in visualizations 608 and 610 may correspond to the current LiDAR point cloud predictions for the vehicle. Thus, for each current LiDAR point, a cost may be determined by comparing each current LiDAR point to the distance function LiDAR value that the current LiDAR point corresponds to or lands on. Thus, for example, current LiDAR point 616 may have a maximum cost while current LiDAR point 618 may have a low cost because point 618 lands close to the center of the points from the LiDAR point cloud in the map data 208. Ultimately, an average or other weighting of each cost from the current LiDAR point may be calculated and the final cost value may be used in addition to the cost values ​​from the elevation and intensity comparisons to update the cost space for the currently sampled pose.

[0053] In some examples, at least one of LiDAR-based cost space sampling 202, cost space aggregation 204, and / or filtering 306 can be performed on a GPU (e.g., a discrete GPU, a virtual GPU, etc.) and / or using one or more parallel processing units. For example, for a camera-based cost space (e.g., FIG. 4A to FIG. 4CAs described above, the detection information and map data 208 projections may be stored as textures in a memory on the GPU or in a memory accessible to the GPU, and the comparison may correspond to a texture lookup performed using the GPU. Similarly, with respect to LiDAR and / or RADAR, the comparison may correspond to a texture lookup. In addition, in some examples, parallel processing may be used to execute two or more cost spaces in parallel—for example, a first cost space corresponding to LiDAR and a second cost space corresponding to RADAR may be generated in parallel using different GPU and / or parallel processing unit resources. For example, individual positioning 226 may be calculated in parallel, reducing the running time of the system for fused positioning. As a result, these processes may be executed more efficiently than when executed only on the CPU.

[0054] See again Figure 2A , after cost space sampling 202 is performed for a single frame or time step and for any number of sensor modalities, cost space aggregation 204 may be performed. Cost space aggregation 204 may be performed separately for each sensor modality - e.g., LiDAR-based cost space aggregation, RADAR-based cost space aggregation, camera-based cost space aggregation, etc. For example, cost space aggregation 204 may aggregate cost spaces calculated for any number of frames (e.g., 25 frames, 80 frames, 100 frames, 300 frames, etc.). To aggregate the cost spaces, each previous cost space that has been calculated may be ego-motion compensated to correspond to the current frame. For example, the rotation and / or translation of the vehicle relative to the current pose of the vehicle and the rotation and / or translation of the vehicle from each previous frame included in the aggregation may be determined, and the cost space values ​​from the previous frame may be carried forward using the rotation and / or translation. In addition to transforming the previous cost space based on ego motion, the cost space may also be transformed so that each cost space corresponds to the origin 212 of the current road segment 214 of the fused HD map. For example, when the vehicle is positioned relative to the first road segment 214, some amount of cost space to be aggregated may have been generated, and when the vehicle is positioned relative to the second road segment 214, some other amount of cost space to be aggregated may have been generated. Thus, the cost space from the first or previous road segment 214 may be transformed so that the cost space values ​​are relative to the second or current road segment 214. Once in the same reference frame corresponding to the current frame and the current road segment 214, the cost space may be aggregated. Thus, and the reference Figure 3B , the self-motion of the vehicle over time can help disambiguate the individual cost spaces so that a total cost space 310 can be generated.

[0055] The total cost space 310 can then undergo filtering 306—e.g., using a Kalman filter or another filter type—to determine an ellipsoid 308 corresponding to the calculated position of the vehicle relative to the current road segment 214. Similar to the description above regarding the transformation for the total cost space 310, the filtered cost space 310 can also undergo a transformation to compensate for ego motion and road segment switching. The ellipsoid 308 can indicate the current position relative to the current road segment 214 of the vehicle, and the process can be repeated at each new frame. The result can be an individual positioning based on the sensor modality for which the ellipsoid 308 is calculated, and multiple ellipsoids 308 can be calculated at each frame—e.g., one ellipsoid for each sensor modality.

[0056] Then, a localization fusion 220 may be performed on the individual localizations 226 to generate a final localization result. Figure 7 , individual positioning 226 can correspond to LiDAR-based positioning 702 (e.g., represented by the ellipsoid and origin in visualization 1300), RADAR-based positioning 704, camera-based positioning 706, other sensor modality positioning (not shown), and / or fused positioning 708. Although only a single positioning of each sensor modality is described herein, this is not intended to be limiting. In some examples, there may be more than one positioning result for different sensor modalities. For example, a vehicle can be positioned relative to a first camera (e.g., a camera facing forward) and can be positioned separately relative to a second camera (e.g., a camera facing backward). In this instance, individual positioning 226 may include a first camera-based positioning and a second camera-based positioning. In some embodiments, fused positioning 708 may correspond to the fusion of individual positioning 226 at the current frame and / or may correspond to one or more previous fused positioning results from one or more previous frames carried forward (e.g., based on own motion) to the current frame. As such, in various embodiments, the fused position fix 708 for the current frame may take into account the individual position fixes 226 and the previous fused position fix results to advance the current position fix state through each frame.

[0057] In order to calculate the fused positioning 708 of the current frame, a consistent / inconsistent analysis can be performed on the individual positioning 226. For example, in some embodiments, a distance threshold can be used to determine the cluster of individual positioning 226, and the cluster with the smallest intra-cluster covariance can be selected for fusion. The individual positioning 226 within the selected cluster can then be averaged or otherwise weighted to determine the fused positioning 708 of the current frame. In some embodiments, a filter (such as a Kalman filter) can be used to generate a fused positioning 708 of clustered individual positioning 226 for the current frame. For example, in the case of using a Kalman filter, the Kalman filter may not handle outliers well, so severe outliers may have an undesirable effect on the final result. As such, the clustering method can help filter out or remove outliers, making the fusion based on the Kalman filter more accurate. In some examples, such as when the previous fusion result is forwarded to the current frame as the individual positioning 226, the fusion result from the previous frame can drift. For example, once the current individual location 226 is sufficiently different from the fused result (e.g., where the fused result can be filtered out of the cluster), the fused result can be reinitialized for the current frame, and then the reinitialized fused result can be carried forward to subsequent frames until a certain amount of drift is detected again.

[0058] In some examples, the fused position 708 can be determined by factoring in each of the individual positions 226. For example, instead of grouping the results into clusters, each individual position 226 can be weighted based on a distance assessment. For example, a covariance can be calculated for the individual positions 226, and the individual position 226 with the highest covariance (e.g., corresponding to the largest outlier) can be weighted less to determine the fused position 708. This can be performed using a robust mean so that outliers do not have an undesirable effect on the fused position 708. For example, the distance of each individual position 226 to the robustified mean can be calculated, and the larger the distance, the smaller the weight that the individual position 226 can have in determining the fused position 708.

[0059] The fused position 708 of the current frame may then be used to position the vehicle relative to the road segment 214 of the fused HD map (represented by the map data 208) and / or relative to a global coordinate system. For example, because the road segment 214 may have a known global position, the position of the vehicle to the road segment 214 may have a corresponding global position result. Furthermore, because the local or relative position to the road segment 214 is more accurate than a global or GNSS position result alone, planning and control of the vehicle may be more reliable and safer than a GNSS-based positioning system alone.

[0060] Return to reference Figure 1In an example of a vehicle, process 100 may include positioning component 102 generating and / or outputting pose data 108 and processing data 110. Pose data 108 may represent an estimated pose (also referred to as a "positioning pose") determined by positioning component 102 based at least on performing one or more of the processes described herein. As described herein, the estimated pose may represent the position of the vehicle (e.g., an x-coordinate position, a y-coordinate position, and / or a z-coordinate position), the orientation of the vehicle (e.g., yaw, pitch, and / or roll), and / or any other position information. Processing data 110 may represent information associated with at least a portion of the processing performed by positioning component 102 in determining the estimated pose. For example, processing data 110 may represent one or more correspondences between one or more features represented by sensor data 104 and one or more features represented by map data 106. As described herein, the correspondence may indicate whether a feature represented by the sensor data 104 matches a feature represented by the map data 106, whether a location of a feature represented by the sensor data 104 matches a location of the same feature represented by the map data 106, a cost associated with matching a feature represented by the sensor data 104 with a feature represented by the map data 106, a cost associated with a pose based on feature matching, and / or any other relationship based at least on feature matching.

[0061] As described herein, in some examples, the positioning performed by the positioning component 102 to determine the estimated pose may include one or more errors based at least on one or more factors. Factors may include, but are not limited to, sensor data 104 not representing a sufficient number of features, sensor data 104 being occluded, an incomplete map of the environment in which the vehicle is navigating, processing problems (e.g., noise), etc. In addition, in some examples, the one or more errors may include a first error between the estimated pose and the actual pose of the vehicle in a first coordinate direction (e.g., x-coordinate direction), a second error between the estimated pose and the actual pose of the vehicle in a second coordinate direction (e.g., y-coordinate direction), a third error between the estimated pose and the actual pose of the vehicle in a third coordinate direction (e.g., z-coordinate direction), a fourth error of a yaw difference between the estimated pose and the actual pose of the vehicle, a fifth error of a roll difference between the estimated pose and the actual pose of the vehicle, a sixth error of a pitch difference between the estimated pose and the actual pose of the vehicle, and the like.

[0062] For example, Figure 8 An example of errors 802(1)-(3) (also referred to as "error 802" in the singular or "errors 802" in the plural) between an actual pose 804 associated with a vehicle and an estimated pose 806 (e.g., a positioning pose) associated with the vehicle is shown in accordance with some embodiments of the present disclosure. Figure 8As shown in the example of , the error 802 may include at least a first error 802(1) associated with a distance in a first direction (e.g., the x direction), a second error 802(2) associated with a distance in a second direction (e.g., the y direction), and a third error 802(3) associated with a yaw angle. However, in other examples, the error 802 may further include an error associated with a distance in a third direction (e.g., the z direction), an error associated with pitch, an error associated with roll, and / or any other error.

[0063] Return to see Figure 1 For example, process 100 may include determining one or more errors associated with the estimated pose and / or one or more uncertainties associated with the one or more errors using error component 112. To determine the errors, error component 112 may use different types of information, such as information associated with the motion of the vehicle, information associated with the positioning process, information associated with one or more previous errors of one or more previously estimated poses, and the like.

[0064] For example, and with respect to information associated with the motion of the vehicle, the error component 112 may determine a true pose associated with the vehicle using a previous pose associated with the vehicle and a portion of the sensor data 104 generated using one or more motion sensors (e.g., one or more inertial measurement unit (IMU) sensors, etc.) (also referred to as "motion data"). In some examples, the pose may be considered a "true" pose based on the assumption that the motion data is accurate and / or substantially accurate for at least a short period of time and / or a short travel distance associated with the vehicle. In some examples, the true pose may include and / or correspond to the actual pose of the vehicle. The error component 112 may then determine one or more differences between the true pose and the estimated pose, such as one or more position differences (e.g., differences in x-coordinate direction, y-coordinate direction, and / or z-coordinate position) and / or one or more orientation differences (e.g., differences in yaw, pitch, and / or roll). In some examples, the positioning component 102 may determine one or more errors (in some examples, referred to as "(one or more) first errors") using the one or more differences.

[0065] For example, Fig. 9An example of using motion associated with a vehicle to determine a true pose associated with a vehicle according to some embodiments of the present disclosure is shown. As shown, the error component 112 can initially determine a previous pose 902 associated with the vehicle. In some examples, the previous pose 902 can correspond to an estimated pose of the vehicle determined by the positioning component 102 during a previous instance of positioning the vehicle and / or an estimated pose of the vehicle for which the error component 112 determined one or more errors. The error component 112 can then determine a motion 904 of the vehicle between a time associated with the previous pose 902 and a current time. As shown, in some examples, the motion 904 can be associated with at least a change in distance 906(1) in a first direction (e.g., an x-direction), a distance 906(2) in a second direction (e.g., a y-direction), and an orientation 906(3) (e.g., a change in yaw angle). However, in other examples, the motion 904 can be associated with one or more distances in one or more additional directions (e.g., a z-direction) and / or one or more additional orientation changes (e.g., a change in pitch, a change in roll, etc.).

[0066] The error component 112 can then determine the true pose 908 associated with the vehicle using the previous pose 902 associated with the vehicle and the motion 904 associated with the vehicle. For example, the error component 112 can determine the true pose 908 by updating the current pose 902 based at least on the motion 904.

[0067] Fig.10 An example of determining one or more errors associated with the motion of a vehicle according to some embodiments of the present invention is shown. For example, the error component 112 can determine the error (one or more) associated with the motion using the following equation:

[0068]

[0069]

[0070] In equations (1)-(2), TP is the true pose 908, LP is the estimated pose 806 (eg, the positioning pose), It is 1002. It is 1004, It is 1006, opt x It is 1008, opt y It is 1010, opt θ Also 1006. Thus, the error determined by the error component 112 may include at least and

[0071] Fig.11An example of using the relative motion of a vehicle and positioning corrections to determine one or more errors associated with the motion of the vehicle according to some embodiments of the present disclosure is shown. As shown in the figure, Fig.11 The actual pose 1102 of the vehicle at the current time (e.g., the actual pose 908), the actual pose 1104 of the vehicle at the previous time, the estimated pose 1106 (e.g., the positioning pose) of the vehicle at the current time (e.g., the estimated pose 806), the estimated pose 1108 of the vehicle at the previous time (e.g., the previous positioning pose), and the corrected pose 1110 of the vehicle at the previous time are shown. In addition, the actual pose 1102 can be determined by adding the relative motion to the actual pose 1104, where the relative motion includes 1112, 1114 and α k 1116. The estimated pose 1106 may be determined by adding the relative motion to the corrected pose 1110. Additionally, the corrected pose 1110 may be determined by adding the positioning correction to the estimated pose 1108, wherein the positioning correction includes 1118, 1120 and β k-1 1122.

[0072] Fig.11 Further shown are additional variables associated with the pose, such as θ k 1124, 1126, 1128, 1130, γ k-1 1132, d k 1134, and l k-1 1136. Thus, the error component 112 may determine the error, such as the error in the x-coordinate direction, using the following equation (which may use a Kalman filter framework): Error in the y-coordinate direction and the error associated with yaw

[0073]

[0074]

[0075] Return to reference Figure 1In some examples, and with respect to the localization process, the error component 112 may compare one or more features represented by the sensor data 104 with one or more features represented by the map data 106. Based at least on the comparison, the error component 112 may determine one or more correspondences between the features represented by the sensor data 104 and the features represented by the map data 106. In some examples, the error component 112 performs the comparison and / or determines the one or more correspondences using one or more techniques similar to the techniques used by the localization component 102 described above when performing localization. In some examples, the correspondences may indicate whether the features represented by the sensor data 104 match the features represented by the map data 106, whether the location of the features represented by the sensor data 104 matches the location of the same features represented by the map data 106, the cost associated with matching the features represented by the sensor data 104 with the features represented by the map data 106, the cost associated with the pose based on the feature matching, and / or any other relationship based at least on the feature matching. Furthermore, in some examples, the error component 112 may determine one or more errors (referred to as "(one or more) second errors" in some examples) using the one or more correspondences.

[0076] The error component 112 can then determine (one or more) final errors associated with the positioning based at least on one or more differences (e.g., (one or more) first errors) between the true pose and the estimated pose (e.g., information associated with the movement of the vehicle); (one or more) correspondences (e.g., (one or more) second errors) between (one or more) features represented by the sensor data 104 and (one or more) features represented by the map data 106 (e.g., information associated with the positioning); (one or more) previous errors associated with one or more previous positionings of the vehicle, and / or any other additional information. In some examples, the error component 112 uses one or more algorithms (such as algorithms associated with a Kalman filter framework) that fuse (one or more) first errors associated with the difference between the true pose and the estimated pose with (one or more) second errors directly observed from the measurement when comparing one or more features represented by the sensor data 104 with one or more features represented by the map data 106 to determine (one or more) final errors. Additionally, in some examples, error component 112 may use similar techniques and / or algorithms to determine the uncertainty associated with the final error(s).

[0077] The process 100 may include the error component 112 outputting error data 114 representing one or more first errors associated with the vehicle motion, one or more second errors associated with the comparison of the sensor data 104 with the map data 106, and / or one or more final errors associated with the positioning. In some examples, the process 100 may also include outputting uncertainty data 116 representing one or more uncertainties associated with the one or more final errors. In addition, as the vehicle continues to navigate around the environment, the process 100 may continue to repeat so that the positioning component 102 continues to locate the vehicle within the environment and the error component 112 continues to determine new errors associated with the new positioning determinations. In some examples, the process 100 may then include performing one or more processes using the error data 114 and / or the uncertainty data 116.

[0078] For example, process 100 may include mode component 118 determining whether one or more final errors meet (e.g., are less than or equal to) one or more error thresholds (e.g., one or more error boundaries), wherein one or more error thresholds are represented by threshold data 120. In some examples, different errors may include different error thresholds. For example, a first error associated with an x-coordinate distance may be associated with a first error threshold (e.g., a first distance threshold), a second error associated with a y-coordinate distance may be associated with a second error threshold (e.g., a second distance threshold), a third error associated with a z-coordinate distance may be associated with a third error threshold (e.g., a third distance threshold), a fourth error associated with yaw may be associated with a fourth error threshold, a fifth error associated with pitch may be associated with a fifth error threshold, and / or a sixth error associated with roll may be associated with a sixth error threshold. In such an example, mode component 118 may compare one or more final errors (e.g., each of the final errors) to a corresponding error threshold.

[0079] In some examples, the mode component 118 can cause the vehicle to operate in a first mode when at least one of the one or more final errors does not satisfy a corresponding error threshold, and / or to operate in a second mode when all final errors satisfy a corresponding error threshold. For example, if the first error does not satisfy (e.g., is greater than) the first error threshold, but the other errors satisfy (e.g., are less than or equal to) their respective error thresholds, the mode component 118 can cause the vehicle to operate in the first mode. However, if the mode component 118 determines that all final errors satisfy (e.g., are less than or equal to) their respective error thresholds, the mode component 118 can cause the vehicle to operate in the second mode. However, in some examples, when two or more of the final errors do not satisfy their respective error thresholds, the mode component 118 can cause the vehicle to operate in the first mode.

[0080] As described herein, in some examples, when operating in a first mode, the vehicle may perform one or more first processes, such as outputting an estimated pose to a user (e.g., positioning a pose), causing one or more other systems of the vehicle to use the estimated pose (e.g., for navigating the vehicle), and / or any other operations. Additionally, when operating in a second mode, the vehicle may perform one or more second processes, such as not outputting an estimated pose to a user, causing one or more other systems to avoid using the estimated pose, and / or any other operations. In such examples, the vehicle may perform such processes because the vehicle may trust the estimated pose when one or more final errors associated with the estimated pose are small, but may not trust the estimated pose when one or more final errors associated with the estimated pose are large.

[0081] FIG. 12A to FIG. 12B An example of determining the error associated with the positioning of a vehicle according to some embodiments of the present disclosure is shown. Fig. 12A In the example of , the localization component 102 and / or the error component 112 can process sensor data (e.g., sensor data 104) representing an image 1202. Based at least on the processing, the localization component 102 and / or the error component 112 can identify a plurality of matches 1204(1)-(6) (also referred to singularly as "match 1204" or plurally as "matches 1204") between features represented by the sensor data and features represented by map data (e.g., map data 106). For example, the localization component 102 and / or the error component 112 can determine a first match 1204(1) associated with a pole, a second match 1204(2) associated with a stop sign, a third match 1204(3) associated with another stop sign, a fourth match 1204(4) associated with a street light, a fifth match 1204(5) associated with another pole, and a sixth match 1204(6) associated with a road marking.

[0082] Based at least on the number of matches 1204, the positioning component 102 can determine an accurate estimated pose associated with the vehicle. As such, and based at least on the number of matches 1204 and the estimated pose being accurate, the error component 112 can determine that at least a first error 1206(1) associated with distance in the x-coordinate direction, a second error 1206(2) associated with distance in the y-coordinate direction, and / or a third error 1206(3) associated with yaw can be small. Thus, the vehicle can operate in the second mode described above.

[0083] exist Fig. 12BIn an example of , localization component 102 and / or error component 112 can process sensor data (e.g., sensor data 104) representing image 1208. Based at least on the processing, localization component 102 and / or error component 112 can not identify any features represented by the sensor data that match any features represented by map data (e.g., map data 106). In some examples, localization component 102 and / or error component 112 can not identify any matches based on the vehicle navigating within an environment that does not include any map features and / or includes few map features, such as within a tunnel. As such, localization component 102 can determine an estimated pose associated with the vehicle that is less accurate than the estimated pose based on the localization component 102. Fig. 12A In addition, and based at least on the number of matches and the fact that the estimated pose is not as accurate as the example of Fig. 12A 1206(1)-(3) of the example of the first error 1210(1) associated with the distance in the x-coordinate direction, the second error 1210(2) associated with the distance in the y-coordinate direction, and / or the third error 1210(3) associated with the yaw that is greater than the errors 1206(1)-(3). Thus, the vehicle can operate in the first mode described above.

[0084] See now Figure 13 to Figure 14 , each block of methods 1300 and 1400 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 implemented by a processor executing instructions stored in a memory. Methods 1300 and 1400 can also be embodied as computer-usable instructions stored on a computer storage medium. Methods 1300 and 1400 can be provided by a stand-alone application, a service or a hosted service (standalone or in combination with other hosted services), or a plug-in for another product, to name a few. In addition, by way of example, reference is made to Figure 1 Methods 1300 and 1400 are described. However, these methods 1300 and 1400 may additionally or alternatively be performed by any one system or any combination of systems, including but not limited to those described herein.

[0085] Fig.13A flow chart of a method 1300 for determining one or more errors associated with positioning according to some embodiments of the present disclosure is shown. At block B1302, the method 1300 may include determining a first pose associated with a machine using a positioning system and based at least on sensor data. For example, the positioning component 102 may use at least a portion of the sensor data 104 (such as sensor data 104 generated using one or more location sensors (e.g., GPS)) to determine an initial pose associated with the machine. The positioning component 102 may then use the sensor data 104 and at least another portion of the map data 106 to refine the initial pose in order to determine a first pose (e.g., an estimated pose, a positioned pose, etc.). In some examples, and as described herein, the positioning component 102 refines the initial pose using at least feature matching.

[0086] At block B1304, the method 1300 may include determining a second pose associated with the machine based at least on the motion data. For example, the error component 112 may use the motion data to determine the motion of the machine, such as a distance traveled and / or a change in orientation in one or more directions. The error component 112 may then determine a second pose (e.g., a true pose, etc.) by moving the machine within the environment from a previous pose based on the motion.

[0087] At block B1306, method 1300 may include determining one or more correspondences between one or more first features represented by the sensor data and one or more second features represented by the map data. For example, error component 112 may compare a first feature represented by sensor data 104 with a second feature represented by map data 106 to determine a correspondence. As described herein, a correspondence may indicate whether a feature represented by sensor data 104 matches a feature represented by map data 106, whether a location of a feature represented by sensor data 104 matches a location of the same feature represented by map data 106, a cost associated with matching a feature represented by sensor data 104 with a feature represented by map data 106, a cost associated with a pose based on feature matching, and / or any other relationship based at least on feature matching.

[0088] At block B1308, method 1300 may include determining one or more differences between the first pose and the second pose. For example, error component 112 may determine one or more differences between the first pose and the second pose. As described herein, in some examples, the one or more differences may include one or more position differences (e.g., differences in x-coordinate direction, y-coordinate direction, and / or z-coordinate position) and / or one or more orientation differences (e.g., differences in yaw, pitch, and / or roll).

[0089] At block B1310, method 1300 may include determining an error associated with the first posture based at least on one or more correspondences and one or more differences. For example, error component 112 may determine an error associated with the first posture based at least on (one or more) correspondences and (one or more) differences. In some examples, error component 112 may determine the error based at least on one or more additional factors, such as previous errors associated with previous postures of the machine. In some examples, error component 112 may determine additional information associated with the error, such as uncertainty associated with the error. In addition, in some examples, error component 112 may perform one or more processes based at least on the error, such as determining an operating mode associated with the machine.

[0090] Fig.14 A flow chart of a method 1400 for determining an operating mode based at least on one or more errors associated with positioning according to some embodiments of the present invention is shown. At block B1402, the method 1400 may include: using a positioning system and based at least on sensor data, determining a pose associated with a machine. For example, the positioning component 102 may use at least a portion of the sensor data 104 (such as sensor data 104 generated using one or more location sensors (e.g., GPS)) to determine an initial pose associated with the machine. The positioning component 102 may then use the sensor data 104 and at least another portion of the map data 106 to refine the initial pose in order to determine a pose (e.g., an estimated pose, a positioning pose, etc.). In some examples, and as described herein, the positioning component 102 refines the initial pose using at least feature matching.

[0091] At block B1404, method 1400 may include determining one or more errors associated with the gesture based at least on the positioning error and the motion error. For example, error component 112 may determine the error associated with the gesture using one or more of the techniques described herein. For example, error component 112 may determine the error(s) based on the positioning error (such as a feature mapping between sensor data 104 and map data) and the motion error (such as one or more differences between a gesture and another gesture determined using the motion data).

[0092] At block B1406, method 1400 may include determining whether the one or more errors satisfy one or more error thresholds. For example, mode component 118 may determine whether the error(s) satisfy the error(s). As described herein, the determination may include determining whether a first error associated with the x-coordinate distance satisfies a first error threshold (e.g., a first distance threshold), whether a second error associated with the y-coordinate distance satisfies a second error threshold (e.g., a second distance threshold), whether a third error associated with the z-coordinate distance satisfies a third error threshold (e.g., a third distance threshold), whether a fourth error associated with the yaw satisfies a fourth error threshold, whether a fifth error associated with the pitch satisfies a fifth error threshold, and / or whether a sixth error associated with the roll satisfies a sixth error threshold.

[0093] If it is determined at block B1406 that the one or more errors do not satisfy the one or more error thresholds, then at block B1408, method 1400 may include causing the machine to operate in a first mode. For example, if mode component 118 determines that the errors do not satisfy one or more error thresholds, mode component 118 may cause the machine to operate in a first mode. As described herein, in the first mode, the machine may not output a pose associated with a position determination and / or the machine may not use the pose to perform one or more operations.

[0094] However, if it is determined at block B1406 that the one or more errors satisfy the one or more error thresholds, then at block B1410, the method 1400 may include causing the machine to operate in a second mode. For example, if the mode component 118 determines that the one or more errors satisfy the one or more error thresholds, the mode component 118 may cause the machine to operate in the second mode. As described herein, in the second mode, the machine may output a gesture associated with the positioning and / or the machine may use the gesture to perform one or more operations.

[0095] Example Autonomous Vehicle

[0096] Fig.15Ais an illustration of an example autonomous vehicle 1500 according to some embodiments of the present disclosure. Autonomous vehicle 1500 (referred to herein alternatively as "vehicle 1500") may include, but is not limited to, a passenger vehicle such as a car, a truck, a bus, an emergency vehicle, a shuttle, an electric or motorized bicycle, a motorcycle, a fire truck, a police car, an ambulance, a boat, an engineering vehicle, an underwater vehicle, a robotic vehicle, a drone, an airplane, a vehicle coupled to a trailer (e.g., a semi-trailer tractor for hauling cargo), and / or another type of vehicle (e.g., a vehicle that is unmanned and / or that can accommodate one or more passengers). Autonomous vehicles are generally described according to the level of automation defined by the National Highway Traffic Safety Administration (NHTSA) of the U.S. Department of Transportation and the Society of Automotive Engineers (SAE) in the “Classification and Definition of Terms Related to Driving Automation Systems for Road Motor Vehicles” (Standard No. J3016-201306 issued on June 15, 2018, Standard No. J3016-201609 issued on September 30, 2016, and previous and future versions of the standard). The vehicle 1500 may be capable of operating at one or more of Levels 3 to 5 of the autonomous driving levels. The vehicle 1500 may be capable of operating at one or more of Levels 1 to 5 of the autonomous driving levels. For example, the vehicle 1500 may be capable of driving assistance (Level 1), partial automation (Level 2), conditional automation (Level 3), high automation (Level 4), and / or full automation (Level 5), depending on the embodiment. The term “autonomy” as used herein may include any and / or all types of autonomy of a vehicle 1500 or other machine, such as fully autonomous, highly autonomous, conditionally autonomous, partially autonomous, providing assisted autonomy, semi-autonomous, primarily autonomous, or other designations.

[0097] The vehicle 1500 may include components such as a chassis, a body, wheels (e.g., 2, 4, 6, 8, 18, etc.), tires, axles, and other components of the vehicle. The vehicle 1500 may include a propulsion system 1550, such as an internal combustion engine, a hybrid power plant, an all-electric engine, and / or another type of propulsion system. The propulsion system 1550 may be connected to a drive train of the vehicle 1500, which may include a transmission, to achieve propulsion of the vehicle 1500. The propulsion system 1550 may be controlled in response to receiving a signal from a throttle / accelerator 1552.

[0098] A steering system 1554, which may include a steering wheel, may be used to steer vehicle 1500 (e.g., along a desired path or route) when propulsion system 1550 is operating (e.g., when the vehicle is in motion). Steering system 1554 may receive signals from steering actuator 1556. For fully automated (Level 5) functionality, a steering wheel may be optional.

[0099] Brake sensor system 1546 may be used to operate vehicle brakes in response to receiving signals from brake actuator 1548 and / or brake sensors.

[0100] May include one or more system on chip (SoC) 1504 ( Fig. 15C ) and / or one or more controllers 1536 of one or more GPUs may provide signals (e.g., representing commands) to one or more components and / or systems of the vehicle 1500. For example, the one or more controllers may send signals to operate vehicle brakes via one or more brake actuators 1548, to operate the steering system 1554 via one or more steering actuators 1556, and to operate the propulsion system 1550 via one or more throttles / accelerators 1552. The one or more controllers 1536 may include one or more onboard (e.g., integrated) computing devices (e.g., supercomputers) that process sensor signals and output operating commands (e.g., signals representing commands) to enable autonomous driving and / or assist a human driver in driving the vehicle 1500. The one or more controllers 1536 may include a first controller 1536 for autonomous driving functions, a second controller 1536 for functional safety functions, a third controller 1536 for artificial intelligence functions (e.g., computer vision), a fourth controller 1536 for infotainment functions, a fifth controller 1536 for redundancy in emergency situations, and / or other controllers. In some examples, a single controller 1536 may handle two or more of the above functions, two or more controllers 1536 may handle a single function, and / or any combination thereof.

[0101] The one or more controllers 1536 may provide signals for controlling one or more components and / or systems of the vehicle 1500 in response to sensor data (e.g., sensor input) received from one or more sensors. The sensor data may be received from, for example and without limitation, a global navigation satellite system ("GNSS") sensor 1558 (e.g., a global positioning system sensor), a RADAR sensor 1560, an ultrasonic sensor 1562, a LiDAR sensor 1564, an inertial measurement unit (IMU) sensor 1566 (e.g., an accelerometer, a gyroscope, a magnetic compass, a magnetometer, etc.), a microphone 1596, a stereo camera 1568, a wide angle camera 1570 (e.g., a fisheye camera), an infrared camera 1572, a surround camera 1574 (e.g., a 360 degree camera), a long-range and / or mid-range camera 1598, a speed sensor 1544 (e.g., for measuring the velocity of the vehicle 1500), a vibration sensor 1542, a steering sensor 1540, a brake sensor (e.g., as part of a brake sensor system 1546), and / or other sensor types.

[0102] One or more of the controllers 1536 may receive inputs (e.g., represented by input data) from the instrument cluster 1532 of the vehicle 1500 and provide outputs (e.g., represented by output data, display data, etc.) via a human machine interface (HMI) display 1534, an audible annunciator, a speaker, and / or via other components of the vehicle 1500. These outputs may include information such as vehicle speed, velocity, time, map data (e.g., Fig. 15C The HMI display 1534 may include information such as a high-definition ("HD") map 1522 of the vehicle 1500, location data (e.g., the location of the vehicle 1500 on the map), directions, locations of other vehicles (e.g., an occupancy grid), information about objects and states of objects as sensed by the controller 1536, and the like. For example, the HMI display 1534 may display information about the presence of one or more objects (e.g., street signs, warning signs, traffic light changes, etc.) and / or information about driving maneuvers that the vehicle has made, is making, or will make (e.g., changing lanes now, leaving 34B in two miles, etc.).

[0103] The vehicle 1500 further includes a network interface 1524 that can communicate via one or more networks using one or more wireless antennas 1526 and / or a modem. For example, the network interface 1524 can be capable of communicating via Long Term Evolution (“LTE”), Wideband Code Division Multiple Access (“WCDMA”), Universal Mobile Telecommunications System (“UMTS”), Global System for Mobile Communications (“GSM”), IMT-CDMA Multi-Carrier (“CDMA2000”), etc. The one or more wireless antennas 1526 can also enable communication between objects in the environment (e.g., vehicles, mobile devices, etc.) using one or more local area networks such as Bluetooth, Bluetooth Low Energy (“LE”), Z-wave, ZigBee, etc. and / or one or more low power wide area networks (LPWANs) such as LoRaWAN, SigFox, etc.

[0104] Fig. 15B According to some embodiments of the present disclosure, Fig.15A 1500. The cameras and respective fields of view are an example embodiment and are not intended to be limiting. For example, additional and / or replaceable cameras may be included and / or the cameras may be located at different locations on vehicle 1500.

[0105] The camera type used for the camera may include, but is not limited to, a digital camera that may be suitable for use with components and / or systems of the vehicle 1500. The camera may operate at an automotive safety integrity level (ASIL) B and / or at another ASIL. The camera type may have any image capture rate, such as 60 frames per second (fps), 920fps, 240fps, etc., depending on the embodiment. The camera may be capable of using a rolling shutter, a global shutter, another type of shutter, or a combination thereof. In some examples, the color filter array may include a red transparent (RCCC) color filter array, a red transparent blue (RCCB) color filter array, a red blue green transparent (RBGC) color filter array, a Foveon X3 color filter array, a Bayer sensor (RGGB) color filter array, a monochrome sensor color filter array, and / or another type of color filter array. In some embodiments, a clear pixel camera such as a camera with an RCCC, RCCB, and / or RBGC color filter array may be used in an effort to improve light sensitivity.

[0106] In some examples, one or more of the cameras can be used to perform advanced driver assistance system (ADAS) functions (e.g., as part of a redundant or fail-safe design). For example, a multi-function monocular camera can be installed to provide functions including lane departure warning, traffic sign assistance, and intelligent headlight control. One or more of the cameras (e.g., all of the cameras) can simultaneously record and provide image data (e.g., video).

[0107] One or more of the cameras may be mounted in a mounting assembly such as a custom designed (three-dimensional ("3D") printed) assembly to cut off stray light and reflections from within the car (e.g., reflections from the dashboard reflected in the windshield mirror) that may interfere with the camera's image data capture capabilities. With respect to the wing mirror mounting assembly, the wing mirror assembly may be custom 3D printed so that the camera mounting plate matches the shape of the wing mirror. In some examples, one or more cameras may be integrated into the wing mirror. For side view cameras, one or more cameras may also be integrated into the four pillars at each corner of the cab.

[0108] A camera (e.g., a front-facing camera) with a field of view that includes a portion of the environment in front of the vehicle 1500 can be used for 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 1536 and / or control SoCs. The front-facing camera can be used to perform many of the same ADAS functions as LiDAR, including emergency braking, pedestrian detection, and collision avoidance. The front-facing camera can also be used for ADAS functions and systems, including lane departure warning ("LDW"), autonomous cruise control ("ACC"), and / or other functions such as traffic sign recognition.

[0109] A variety of cameras may be used in the front-facing configuration, including, for example, a monocular camera platform including a complementary metal oxide semiconductor (CMOS) color imager. Another example may be a wide-angle camera 1570, which may be used to sense objects (e.g., pedestrians, intersection traffic, or bicycles) entering the field of view from the periphery. Fig. 15B Only one wide-angle camera is shown in the figure, but there may be any number (including zero) of wide-angle cameras 1570 on the vehicle 1500. In addition, any number of long-range cameras 1598 (e.g., a long-view stereo camera pair) may be used for depth-based object detection, especially for objects for which a neural network has not been trained. Long-range cameras 1598 may also be used for object detection and classification and basic object tracking.

[0110] Any number of stereo cameras 1568 may also be included in the front configuration. In at least one embodiment, one or more stereo cameras 1568 may include an integrated control unit including a scalable processing unit that may provide a multi-core microprocessor and programmable logic ("FPGA") with an integrated controller area network ("CAN") or Ethernet interface on a single chip. Such a unit can be used to generate a 3D map of the vehicle environment, including distance estimates for all points in the image. Alternative stereo cameras 1568 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 autonomous emergency braking and lane departure warning functions. Other types of stereo cameras 1568 may be used in addition to or alternatively to those described herein.

[0111] A camera (e.g., a side view camera) having a field of view of a portion of the environment including the sides of the vehicle 1500 can be used for surround viewing, providing information used to create and update the occupancy grid and generate side impact collision warnings. Fig. 15BFour surround cameras 1574 (shown in FIG. 1 ) may be positioned on the vehicle 1500. The surround cameras 1574 may include a wide-angle camera 1570, a fisheye camera, a 360-degree camera, and / or the like. For example, four fisheye cameras may be positioned in front, at the rear, and on the sides of the vehicle. In an alternative arrangement, the vehicle may use three surround cameras 1574 (e.g., left, right, and rear), and may utilize one or more other cameras (e.g., a forward-facing camera) as a fourth surround camera.

[0112] A camera having a field of view that includes a portion of the environment behind the vehicle 1500 (e.g., a rear view camera) can be used to assist with parking, surround view, rear collision warning, and creating and updating an occupancy grid. A variety of cameras can be used, including but not limited to cameras that are also suitable as front-facing cameras as described herein (e.g., long-range and / or mid-range cameras 1598, stereo cameras 1568, infrared cameras 1572, etc.).

[0113] Fig. 15C According to some embodiments of the present disclosure, Fig.15A Block diagram of an example system architecture of an example autonomous vehicle 1500. It should be understood that this arrangement and other arrangements described herein are set forth merely as examples. Other arrangements and elements (e.g., machines, interfaces, functions, sequences, functional groupings, etc.) may be used in addition to or in place of those shown, and some elements may be omitted entirely. Further, many of the elements described herein are functional entities that may be implemented as discrete or distributed components or in combination with other components, and in any appropriate combination and location. The various functions described herein as being performed by an entity may be implemented by hardware, firmware, and / or software. For example, the various functions may be implemented by a processor executing instructions stored in a memory.

[0114] Fig. 15C Each of the components, features, and systems of vehicle 1500 is illustrated as being connected via bus 1502. Bus 1502 may include a controller area network (CAN) data interface (alternatively, referred to herein as a "CAN bus"). CAN may be a network internal to vehicle 1500 that assists in controlling various features and functions of vehicle 1500, such as actuation of brakes, acceleration, braking, steering, windshield wipers, and the like. The CAN bus may be configured to have tens or even hundreds of nodes, each with its own unique identifier (e.g., CAN ID). The CAN bus may be read to find steering wheel angle, ground speed, engine speed per minute (RPM), button position, and / or other vehicle status indicators. The CAN bus may be ASIL B compliant.

[0115] Although bus 1502 is described as a CAN bus here, this is not intended to be restrictive. For example, in addition to or alternatively to the CAN bus, FlexRay and / or Ethernet can be used. In addition, although bus 1502 is represented by a single line, this is not intended to be restrictive. For example, there can be any number of buses 1502, which may include one or more CAN buses, one or more FlexRay buses, one or more Ethernet buses, and / or one or more other types of buses using different protocols. In some examples, two or more buses 1502 can be used to perform different functions, and / or can be used for redundancy. For example, a first bus 1502 can be used for a collision avoidance function, and a second bus 1502 can be used for drive control. In any example, each bus 1502 can communicate with any component of the vehicle 1500, and two or more buses 1502 can communicate with the same component. In some examples, each SoC 1504, each controller 1536, and / or each computer within the vehicle may have access to the same input data (e.g., input from sensors of the vehicle 1500) and may be connected to a common bus such as a CAN bus.

[0116] The vehicle 1500 may include one or more controllers 1536, such as those described herein. Fig.15A Those controllers described. Controller 1536 can be used for a variety of functions. Controller 1536 can be coupled to any other different components and systems of vehicle 1500, and can be used for control of vehicle 1500, artificial intelligence of vehicle 1500, infotainment for vehicle 1500, and / or the like.

[0117] The vehicle 1500 may include one or more system on chip (SoC) 1504. The SoC 1504 may include a CPU 1506, a GPU 1508, a processor 1510, a cache 1512, an accelerator 1514, a data store 1516, and / or other components and features not shown. The SoC 1504 may be used to control the vehicle 1500 in a variety of platforms and systems. For example, one or more SoCs 1504 may be combined with an HD map 1522 in a system (e.g., a system of the vehicle 1500), and the HD map may be downloaded from one or more servers (e.g., a server) via a network interface 1524. Fig.15D one or more servers 1578) to obtain map refreshes and / or updates.

[0118] CPU 1506 may include a CPU cluster or CPU complex (alternatively, referred to herein as "CCPLEX"). CPU 1506 may include multiple cores and / or L2 caches. For example, in some embodiments, CPU 1506 may include eight cores in a coherent multiprocessor configuration. In some embodiments, CPU 1506 may include four dual-core clusters, each of which has a dedicated L2 cache (e.g., 2MB L2 cache). CPU 1506 (e.g., CCPLEX) may be configured to support simultaneous cluster operations so that any combination of clusters of CPU 1506 can be active at any given time.

[0119] CPU 1506 may implement power management capabilities including one or more of the following features: each hardware block may be automatically clock gated when idle to save dynamic power; each core clock may be gated when the core is not actively executing instructions due to the execution of WFI / WFE instructions; each core may be independently power gated; each core cluster may be independently clock gated when all cores are clock gated or power gated; and / or each core cluster may be independently power gated when all cores are power gated. CPU 1506 may further implement an enhanced algorithm for managing power states, in which allowed power states and expected wake-up times are specified, and hardware / microcode determines the optimal power state to enter for the core, cluster, and CCPLEX. The processing core may support a simplified power state entry sequence in software, with the work being offloaded to the microcode.

[0120] GPU 1508 may include an integrated GPU (alternatively, referred to herein as an "iGPU"). GPU 1508 may be programmable and efficient for parallel workloads. In some examples, GPU 1508 may use an enhanced tensor instruction set. GPU 1508 may include one or more streaming microprocessors, each of which may include an L1 cache (e.g., an L1 cache with at least 96KB storage capacity), and two or more of these streaming microprocessors may share an L2 cache (e.g., an L2 cache with 512KB storage capacity). In some embodiments, GPU 1508 may include at least eight streaming microprocessors. GPU 1508 may use a computing application programming interface (API). In addition, GPU 1508 may use one or more parallel computing platforms and / or programming models (e.g., NVIDIA's CUDA).

[0121] In the case of automotive and embedded use, GPU 1508 can be power optimized to achieve optimal performance. For example, GPU 1508 can be manufactured on fin field effect transistors (FinFETs). However, this is not intended to be limiting, and GPU 1508 can be manufactured using other semiconductor manufacturing processes. Each streaming microprocessor can merge several mixed precision processing cores divided into multiple blocks. For example and without limitation, 64 PF32 cores and 32 PF64 cores can be divided into four processing blocks. In such an example, each processing block can be allocated 16 FP32 cores, 8 FP64 cores, 16 INT32 cores, two mixed precision NVIDIA tensor cores for deep learning matrix arithmetic, L0 instruction cache, warp scheduler, dispatch unit and / or 64KB register file. In addition, the streaming microprocessor may include independent parallel integer and floating point data paths to provide efficient execution of workloads using a mix of computation and addressing computations. The streaming microprocessor may include independent thread scheduling capabilities to allow for finer-grained synchronization and cooperation between parallel threads. The streaming microprocessor may include a combined L1 data cache and shared memory unit to improve performance while simplifying programming.

[0122] GPU 1508 can include high bandwidth memory (HBM) and / or a 16GB HBM2 memory subsystem that provides a peak memory bandwidth of approximately 900 GB / s in some examples. In some examples, synchronous graphics random access memory (SGRAM), such as fifth generation graphics double data rate synchronous random access memory (GDDR5), can be used in addition to or in lieu of HBM memory.

[0123] The GPU 1508 may include unified memory technology that includes access counters to allow memory pages to be more accurately migrated to the processor that accesses them most frequently, thereby improving the efficiency of memory ranges shared between processors. In some examples, address translation service (ATS) support may be used to allow the GPU 1508 to directly access the CPU 1506 page table. In such an example, when the GPU 1508 memory management unit (MMU) experiences a miss, an address translation request may be transmitted to the CPU 1506. In response, the CPU 1506 may look up the virtual-physical mapping for the address in its page table and transmit the translation back to the GPU 1508. In this way, the unified memory technology may allow a single unified virtual address space to be used for memory of both the CPU 1506 and the GPU 1508, thereby simplifying GPU 1508 programming and porting applications to the GPU 1508.

[0124] In addition, GPU 1508 may include access counters that can track how often GPU 1508 accesses the memory of other processors. The access counters can help ensure that memory pages are moved to the physical memory of the processor that accesses them most frequently.

[0125] SoC 1504 may include any number of caches 1512, including those described herein. For example, cache 1512 may include an L3 cache available to both CPU 1506 and GPU 1508 (e.g., connected to both CPU 1506 and GPU 1508). Cache 1512 may include a write-back cache that may track the state of a line, for example, by using a cache coherence protocol (e.g., MEI, MESI, MSI, etc.). Depending on the embodiment, the L3 cache may include 4MB or more, although smaller cache sizes may also be used.

[0126] The SoC 1504 may include an arithmetic logic unit (ALU) that may be used to perform processing for any of a variety of tasks or operations with respect to the vehicle 1500 - such as processing a DNN. In addition, the SoC 1504 may include a floating point unit (FPU) - or other math coprocessor or digital coprocessor type - for performing mathematical operations within the system. For example, the SoC 104 may include one or more FPUs integrated as execution units within the CPU 1506 and / or GPU 1508.

[0127] SoC 1504 may include one or more accelerators 1514 (e.g., hardware accelerators, software accelerators, or a combination thereof). For example, SoC 1504 may include a hardware acceleration cluster, which may include optimized hardware accelerators and / or large on-chip memory. The large on-chip memory (e.g., 4MB SRAM) may enable the hardware acceleration cluster to accelerate neural networks and other calculations. The hardware acceleration cluster may be used to supplement GPU 1508 and offload some tasks of GPU 1508 (e.g., free up more cycles of GPU 1508 for performing other tasks). As an example, accelerator 1514 may be used for targeted workloads (e.g., perception, convolutional neural networks (CNNs), etc.) that are sufficiently stable to easily control acceleration. When used herein, the term "CNN" may include all types of CNNs, including region-based or regional convolutional neural networks (RCNNs) and fast RCNNs (e.g., for object detection).

[0128] Accelerator 1514 (e.g., a hardware acceleration cluster) may include a deep learning accelerator (DLA). DLA may include one or more tensor processing units (TPUs) that may be configured to provide an additional 10 trillion operations per second for deep learning applications and reasoning. TPU may be an accelerator configured to perform image processing functions (e.g., for CNN, RCNN, etc.) and optimized for performing image processing functions. DLA may be further optimized for a specific set of neural network types and floating point operations and reasoning. The design of DLA may provide higher performance per millimeter than a general-purpose GPU, and far exceeds the performance of a CPU. TPU may perform several functions, including a single instance convolution function, support, for example, INT8, INT16, and FP16 data types for both features and weights, and post-processor functions.

[0129] The DLA can quickly and efficiently execute neural networks, particularly CNNs, on processed or unprocessed data for any of a wide variety of functions, such as, but not limited to: a CNN for object recognition and detection using data from a camera sensor; a CNN for distance estimation using data from a camera sensor; a CNN for emergency vehicle detection and identification and detection using data from a microphone; a CNN for facial recognition and vehicle owner identification using data from a camera sensor; and / or a CNN for safety and / or security-related events.

[0130] The DLA can perform any function of the GPU 1508, and by using an inference accelerator, for example, the designer can target any function to either the DLA or the GPU 1508. For example, the designer can focus the processing of CNNs and floating point operations on the DLA, and leave other functions to the GPU 1508 and / or other accelerators 1514.

[0131] Accelerator 1514 (e.g., a hardware acceleration cluster) may include a programmable vision accelerator (PVA), which may be alternatively referred to herein as a computer vision accelerator. The PVA may be designed and configured to accelerate computer vision algorithms for advanced driver assistance systems (ADAS), autonomous driving, and / or augmented reality (AR) and / or virtual reality (VR) applications. The PVA may provide a balance between performance and flexibility. For example, each PVA may include, for example and without limitation, any number of reduced instruction set computer (RISC) cores, direct memory access (DMA), and / or any number of vector processors.

[0132] The RISC core can interact with an image sensor (e.g., an image sensor of any camera described herein), an image signal processor, and / or the like. Each of these RISC cores can include any number of memories. Depending on the embodiment, the RISC core can use any of a number of protocols. In some examples, the RISC core can execute a real-time operating system (RTOS). The RISC core can be implemented using one or more integrated circuit devices, application specific integrated circuits (ASICs), and / or storage devices. For example, the RISC core can include an instruction cache and / or a tightly coupled RAM.

[0133] The DMA may enable components of the PVA to access system memory independently of the CPU 1506. The DMA may support any number of features used to provide optimizations for the PVA, including, but not limited to, support for multi-dimensional addressing and / or circular addressing. In some examples, the DMA may support addressing in up to six or more dimensions, which may include block width, block height, block depth, horizontal block stepping, vertical block stepping, and / or depth stepping.

[0134] The vector processor can be a programmable processor that can be designed to efficiently and flexibly perform programming for computer vision algorithms and provide signal processing capabilities. In some examples, the PVA can include a PVA core and two vector processing subsystem partitions. The PVA core can include a processor subsystem, one or more DMA engines (e.g., two DMA engines), and / or other peripherals. The vector processing subsystem can operate as the main processing engine of the PVA, and can include a vector processing unit (VPU), an instruction cache, and / or a vector memory (e.g., VMEM). The VPU core can include a digital signal processor, such as, for example, a single instruction multiple data (SIMD), a very long instruction word (VLIW) digital signal processor. The combination of SIMD and VLIW can enhance throughput and rate.

[0135] Each of the vector processors may include an instruction cache and may be coupled to a dedicated memory. As a result, in some examples, each of the vector processors may be configured to execute independently of the other vector processors. In other examples, the vector processors included in a particular PVA may be configured to employ data parallelism. For example, in some embodiments, multiple vector processors included in a single PVA may execute the same computer vision algorithm, but on different regions of an image. In other examples, the vector processors included in a particular PVA may execute different computer vision algorithms simultaneously on the same image, or even execute different algorithms on sequence images or portions of an image. Among other things, any number of PVAs may be included in a hardware acceleration cluster, and any number of vector processors may be included in each of these PVAs. In addition, the PVA may include additional error correction code (ECC) memory to enhance overall system security.

[0136] The accelerator 1514 (e.g., a hardware acceleration cluster) may include an on-chip computer vision network and SRAM to provide high bandwidth, low latency SRAM for the accelerator 1514. In some examples, the on-chip memory may include at least 4MB of SRAM consisting of, for example and not limited to, eight field-configurable memory blocks that can be accessed by both the PVA and the DLA. Each pair of memory blocks may include an advanced peripheral bus (APB) interface, configuration circuitry, a controller, and a multiplexer. Any type of memory may be used. The PVA and DLA may access the memory via a backbone that provides high-speed memory access to the PVA and DLA. The backbone may include an on-chip computer vision network that interconnects the PVA and DLA to the memory (e.g., using APB).

[0137] The on-chip computer vision network may include an interface that determines that both the PVA and DLA provide ready and valid signals before transmitting any control signals / addresses / data. Such an interface may provide separate phases and separate channels for transmitting control signals / addresses / data, as well as burst communication for continuous data transmission. This type of interface may comply with ISO 26262 or IEC 61508 standards, but other standards and protocols may also be used.

[0138] In some examples, SoC 1504 may include a real-time ray tracing hardware accelerator such as described in U.S. Patent Application No. 16 / 101,232 filed on August 10, 2018. The real-time ray tracing hardware accelerator may be used to quickly and efficiently determine the position and range of objects (e.g., within a world model) in order to generate real-time visualization simulations for RADAR signal interpretation, for sound propagation synthesis and / or analysis, for SONAR system simulations, for general wave propagation simulations, for comparison with LiDAR data for the purpose of positioning and / or other functions, and / or for other purposes. In some embodiments, one or more tree traversal units (TTUs) may be used to perform one or more ray tracing related operations.

[0139] The accelerator 1514 (e.g., a hardware accelerator cluster) has a wide range of uses in autonomous driving. The PVA can be a programmable visual accelerator that can be used for key processing stages in ADAS and autonomous vehicles. The capabilities of the PVA are a good match for algorithmic domains that require predictable processing, low power, and low latency. In other words, the PVA performs well on semi-dense or dense rule computations, and even on small data sets that require predictable runtimes with low latency and low power. Therefore, in the context of a platform for autonomous vehicles, the PVA is designed to run classic computer vision algorithms because they are efficient at object detection and integer math.

[0140] For example, according to one embodiment of the technology, PVA is used to perform computer stereo vision. In some examples, algorithms based on semi-global matching can be used, but this is not intended to be limiting. Many applications for level 3-5 autonomous driving require instant motion estimation / stereo matching (e.g., structure from motion, pedestrian recognition, lane detection, etc.). PVA can perform computer stereo vision functions on input from two monocular cameras.

[0141] In some examples, PVA can be used to perform dense optical flow. According to processing raw RADAR data (e.g., using 4D fast Fourier transform) to provide processed RADAR. In other examples, PVA is used for time-of-flight depth processing, which is, for example, by processing raw time-of-flight data to provide processed time-of-flight data.

[0142] 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 as providing a relative "weight" of each detection compared to other detections. The confidence value enables the system to make further decisions about which detections should be considered true positive detections rather than false positive detections. For example, the system can set a threshold for confidence and only consider detections that exceed the threshold as true positive detections. In an automatic emergency braking (AEB) system, a false positive detection can cause the vehicle to automatically perform emergency braking, which is obviously undesirable. Therefore, only the most confident detections should be considered as triggers for AEB. DLA can run a neural network for regressing confidence values. The neural network may take as its input at least some subset of parameters, such as bounding box dimensions, a ground plane estimate obtained (e.g., from another subsystem), an inertial measurement unit (IMU) sensor 1566 output related to the orientation and distance of the vehicle 1500, a 3D position estimate of an object obtained from the neural network and / or other sensors (e.g., a LiDAR sensor 1564 or a RADAR sensor 1560), etc.

[0143] SoC 1504 may include one or more data stores 1516 (e.g., memory). Data store 1516 may be on-chip memory of SoC 1504 that may store neural networks to be executed on the GPU and / or DLA. In some examples, data store 1516 may be large enough to store multiple instances of a neural network for redundancy and safety. Data store 1512 may include an L2 or L3 cache 1512. References to data store 1516 may include references to memory associated with a PVA, DLA, and / or other accelerator 1514 as described herein.

[0144] SoC 1504 may include one or more processors 1510 (e.g., embedded processors). Processor 1510 may include a boot and power management processor, which may be a dedicated processor and subsystem for handling boot power and management functions and related safety implementations. The boot and power management processor may be part of the SoC 1504 boot sequence and may provide runtime power management services. The boot power and management processor may provide clock and voltage programming, auxiliary system low power state transitions, SoC 1504 thermal and temperature sensor management, and / or SoC 1504 power state management. Each temperature sensor may be implemented as a ring oscillator whose output frequency is proportional to the temperature, and SoC 1504 may use the ring oscillator to detect the temperature of CPU 1506, GPU 1508, and / or accelerator 1514. If it is determined that the temperature exceeds a threshold, the boot and power management processor may enter a temperature fault routine and place SoC 1504 in a lower power state and / or place vehicle 1500 in a driver safety parking mode (e.g., parking vehicle 1500 safely).

[0145] Processor 1510 may further include a set of embedded processors that may be used as an audio processing engine. The audio processing engine may be an audio subsystem that allows for full hardware support for multi-channel audio through multiple interfaces and a wide range of flexible audio I / O interfaces. In some examples, the audio processing engine is a dedicated processor core having a digital signal processor with dedicated RAM.

[0146] Processor 1510 may further include an always-on processor engine that may provide the necessary hardware features to support low-power sensor management and wake-up use cases. The always-on processor engine may include a processor core, tightly coupled RAM, supporting peripherals (e.g., timers and interrupt controllers), various I / O controller peripherals, and routing logic.

[0147] The processor 1510 may further include a safety cluster engine, which includes a dedicated processor subsystem that handles safety management of automotive applications. The safety cluster engine may include two or more processor cores, tightly coupled RAM, supporting peripherals (e.g., timers, interrupt controllers, etc.), and / or routing logic. In safety mode, the two or more cores may operate in lockstep mode and act as a single core with comparison logic to detect any differences between their operations.

[0148] Processor 1510 may further include a real-time camera engine, which may include a dedicated processor subsystem for handling real-time camera management.

[0149] Processor 1510 may further include a high dynamic range signal processor, which may include an image signal processor, which is a hardware engine that is part of the camera processing pipeline.

[0150] The processor 1510 may include a video image compositer, which may be a processing block (e.g., implemented on a microprocessor) that implements video post-processing functions required for a video playback application to produce a final image for a player window. The video image compositer may perform lens distortion correction for the wide-angle camera 1570, the surround camera 1574, and / or for an in-cab surveillance camera sensor. The in-cab surveillance camera sensor is preferably monitored by a neural network running on another instance of the advanced SoC, configured to recognize in-cab events and respond accordingly. The in-cab system may perform lip reading to activate mobile phone service and place a call, dictate an email, change a vehicle destination, activate or change the vehicle's infotainment system and settings, or provide voice-activated web surfing. Certain functions are only available to the driver when the vehicle is operating in autonomous mode and are disabled in other circumstances.

[0151] The video image compositer may include enhanced temporal noise reduction for both spatial and temporal noise reduction. For example, in the case of motion in the video, the noise reduction appropriately weights the spatial information and reduces the weight of information provided by adjacent frames. In the case where an image or portion of an image does not include motion, the temporal noise reduction performed by the video image compositer may use information from previous images to reduce noise in the current image.

[0152] The video image compositor may also be configured to perform stereoscopic rectification on the input stereoscopic footage frames. The video image compositor may further be used for user interface composition when the operating system desktop is in use and the GPU 1508 does not need to continuously render new surfaces. Even when the GPU 1508 is powered on and performing 3D rendering, the video image compositor may be used to offload the GPU 1508 to improve performance and responsiveness.

[0153] SoC 1504 may further include a mobile industry processor interface (MIPI) camera serial interface for receiving video and input from a camera, a high-speed interface, and / or a video input block that may be used for camera and related pixel input functions. SoC 1504 may further include an input / output controller that may be controlled by software and may be used to receive I / O signals that are not committed to a specific role.

[0154] SoC 1504 may further include a wide range of peripheral interfaces to enable communication with peripherals, audio codecs, power management and / or other devices. SoC 1504 may be used to process data from cameras (connected via Gigabit multimedia serial link and Ethernet), sensors (e.g., LiDAR sensor 1564, RADAR sensor 1560, etc., which may be connected via Ethernet), data from bus 1502 (e.g., speed of vehicle 1500, steering wheel position, etc.), data from GNSS sensor 1558 (connected via Ethernet or CAN bus). SoC 1504 may further include dedicated high-performance mass storage controllers, which may include their own DMA engines, and which may be used to free up CPU 1506 from routine data management tasks.

[0155] SoC 1504 can be an end-to-end platform with a flexible architecture that spans automation levels 3-5, providing a comprehensive functional safety architecture that leverages and efficiently uses computer vision and ADAS technologies to achieve diversity and redundancy, together with deep learning tools to provide a platform for a flexible and reliable driving software stack. SoC 1504 can be faster, more reliable, and even more energy efficient and space efficient than conventional systems. For example, when combined with CPU 1506, GPU 1508, and data storage 1516, accelerator 1514 can provide a fast and efficient platform for level 3-5 autonomous vehicles.

[0156] The technology thus provides capabilities and functionality that cannot be achieved by conventional systems. For example, computer vision algorithms can be executed on CPUs, which can be configured using high-level programming languages ​​such as the C programming language to execute a wide variety of processing algorithms across a wide variety of visual data. However, CPUs often fail to meet the performance requirements of many computer vision applications, such as those related to, for example, execution time and power consumption. In particular, many CPUs are unable to execute complex object detection algorithms in real time, which is a requirement for in-vehicle ADAS applications and a requirement for practical Level 3-5 autonomous vehicles.

[0157] In contrast to conventional systems, by providing a CPU complex, a GPU complex, and a hardware acceleration cluster, the techniques described herein allow multiple neural networks to be executed simultaneously and / or sequentially, and the results to be combined together to achieve Level 3-5 autonomous driving functions. For example, a CNN executed on a DLA or dGPU (e.g., GPU 1520) may include text and word recognition, allowing the supercomputer to read and understand traffic signs, including signs for which a neural network has not been specifically trained. The DLA may further include a neural network capable of recognizing, interpreting, and providing semantic understanding of the signs, and passing that semantic understanding to a path planning module running on the CPU complex.

[0158] As another example, as required for Level 3, 4, or 5 driving, multiple neural networks may be running simultaneously. For example, a warning sign consisting of "Caution: Flashing Lights Indicate Icing Conditions" along with electric lights may be interpreted by several neural networks independently or collectively. The sign itself may be recognized as a traffic sign by a deployed first neural network (e.g., a trained neural network), and the text "Flashing Lights Indicate Icing Conditions" may be interpreted by a deployed second neural network that informs the vehicle's path planning software (preferably executing on a CPU complex) that icing conditions exist when the flashing lights are detected. The flashing lights may be recognized by operating a deployed third neural network over multiple frames that informs the vehicle's path planning software of the presence (or absence) of the flashing lights. All three neural networks may be running simultaneously, for example, within a DLA and / or on GPU 1508.

[0159] In some examples, a CNN for facial recognition and owner recognition can use data from the camera sensor to identify the presence of an authorized driver and / or owner of the vehicle 1500. The always-on sensor processing engine can be used to unlock the vehicle and turn on the lights when the owner approaches the driver's door, and in security mode, disable the vehicle when the owner leaves the vehicle. In this way, the SoC 1504 provides security against theft and / or carjacking.

[0160] In another example, a CNN for emergency vehicle detection and identification can use data from microphone 1596 to detect and identify emergency vehicle sirens. In contrast to conventional systems that use general classifiers to detect sirens and manually extract features, SoC 1504 uses CNN to classify environmental and urban sounds and to classify visual data. In a preferred embodiment, the CNN running on the DLA is trained to identify the relative closing rate of emergency vehicles (e.g., by using the Doppler effect). The CNN can also be trained to identify emergency vehicles specific to the local area in which the vehicle is operating, as identified by the GNSS sensor 1558. Thus, for example, when operating in Europe, the CNN will seek to detect European sirens, and when in the United States, the CNN will seek to identify only North American sirens. Once an emergency vehicle is detected, with the assistance of the ultrasonic sensor 1562, the control program can be used to execute emergency vehicle safety routines to slow the vehicle, drive to the side of the road, stop the vehicle, and / or idle the vehicle until the emergency vehicle passes.

[0161] The vehicle may include a CPU 1518 (e.g., a discrete CPU or dCPU) that may be coupled to the SoC 1504 via a high-speed interconnect (e.g., PCIe). The CPU 1518 may include, for example, an X86 processor. The CPU 1518 may be used to perform any of a variety of functions, including, for example, arbitrating potentially inconsistent results between ADAS sensors and the SoC 1504, and / or monitoring the status and health of the controller 1536 and / or the infotainment SoC 1530.

[0162] The vehicle 1500 may include a GPU 1520 (e.g., a discrete GPU or dGPU) that may be coupled to the SoC 1504 via a high-speed interconnect (e.g., NVIDIA's NVLINK). The GPU 1520 may provide additional artificial intelligence functionality, for example, by executing redundant and / or different neural networks, and may be used to train and / or update the neural network based on inputs (e.g., sensor data) from sensors of the vehicle 1500.

[0163] The vehicle 1500 may further include a network interface 1524, which may include one or more wireless antennas 1526 (e.g., one or more wireless antennas for different communication protocols, such as a cellular antenna, a Bluetooth antenna, etc.). The network interface 1524 can be used to enable wireless connections to the cloud (e.g., with a server 1578 and / or other network devices), to other vehicles, and / or to computing devices (e.g., a passenger's client device) via the Internet. In order to communicate with other vehicles, a direct link can be established between the two vehicles, and / or an indirect link can be established (e.g., across a network and through the Internet). The direct link can be provided using a vehicle-to-vehicle communication link. The vehicle-to-vehicle communication link can provide the vehicle 1500 with information about vehicles approaching the vehicle 1500 (e.g., vehicles in front of, to the side of, and / or behind the vehicle 1500). This functionality can be part of the cooperative adaptive cruise control functionality of the vehicle 1500.

[0164] The network interface 1524 may include a SoC that provides modulation and demodulation functions and enables the controller 1536 to communicate over a wireless network. The network interface 1524 may include a radio frequency front end for up-conversion from baseband to radio frequency and down-conversion from radio frequency to baseband. The frequency conversion may be performed by a known process and / or may be performed using a super-heterodyne process. In some examples, the radio frequency front end function may be provided by a separate chip. The network interface may include a wireless function for communicating via LTE, WCDMA, UMTS, GSM, CDMA2000, Bluetooth, Bluetooth LE, Wi-Fi, Z-wave, ZigBee, LoRaWAN, and / or other wireless protocols.

[0165] The vehicle 1500 may further include data storage 1528, which may include off-chip storage (e.g., outside the SoC 1504). The data storage 1528 may include one or more storage elements, including RAM, SRAM, DRAM, VRAM, flash memory, hard disk, and / or other components and / or devices that can store at least one bit of data.

[0166] The vehicle 1500 may further include a GNSS sensor 1558. The GNSS sensor 1558 (e.g., GPS, assisted GPS sensor, differential GPS (DGPS) sensor, etc.) is used to assist in mapping, perception, occupancy grid generation, and / or path planning functions. Any number of GNSS sensors 1558 may be used, including, for example and without limitation, GPS using a USB connector with an Ethernet to serial (RS-232) bridge.

[0167] The vehicle 1500 may further include a RADAR sensor 1560. The RADAR sensor 1560 may be used by the vehicle 1500 for remote vehicle detection even in darkness and / or inclement weather conditions. The RADAR functional safety level may be ASILB. The RADAR sensor 1560 may use CAN and / or bus 1502 (e.g., to transmit data generated by the RADAR sensor 1560) for control and access to object tracking data, accessing Ethernet in some examples to access raw data. A variety of RADAR sensor types may be used. For example and without limitation, the RADAR sensor 1560 may be suitable for front, rear, and side RADAR use. In some examples, a pulse Doppler RADAR sensor is used.

[0168] The RADAR sensor 1560 may include different configurations, such as long-range with a narrow field of view, short-range with a wide field of view, short-range side coverage, and the like. In some examples, the long-range RADAR may be used for adaptive cruise control functions. The long-range RADAR system may provide a wide field of view (e.g., within a range of 250m) achieved by two or more independent scans. The RADAR sensor 1560 may help distinguish between static and moving objects and may be used by the ADAS system for emergency braking assistance and forward collision warnings. The long-range RADAR sensor may include a single-station multimode RADAR with multiple (e.g., six or more) fixed RADAR antennas and high-speed CAN and FlexRay interfaces. In an example with six antennas, the central four antennas may create a focused beam pattern designed to record the surroundings of the vehicle 1500 at a higher rate with minimal traffic interference from adjacent lanes. The other two antennas may expand the field of view, making it possible to quickly detect vehicles entering or leaving the lane of the vehicle 1500.

[0169] As an example, a medium-range RADAR system may include a range of up to 960m (front) or 80m (rear) and a field of view of up to 42 degrees (front) or 1550 degrees (rear). A short-range RADAR system may include, but is not limited to, a RADAR sensor designed to be mounted on both ends of a rear bumper. When mounted on both ends of a rear bumper, such a RADAR sensor system may create two beams that continuously monitor the blind spots behind and beside the vehicle.

[0170] Short-range RADAR systems can be used in ADAS systems for blind spot detection and / or lane change assistance.

[0171] The vehicle 1500 may further include ultrasonic sensors 1562. The ultrasonic sensors 1562, which may be placed on the front, rear, and / or sides of the vehicle 1500, may be used for parking assistance and / or creating and updating an occupancy grid. A variety of ultrasonic sensors 1562 may be used, and different ultrasonic sensors 1562 may be used for different detection ranges (e.g., 2.5 m, 4 m). The ultrasonic sensors 1562 may operate at a functional safety level of ASIL B.

[0172] The vehicle 1500 may include a LiDAR sensor 1564. The LiDAR sensor 1564 may be used for object and pedestrian detection, emergency braking, collision avoidance, and / or other functions. The LiDAR sensor 1564 may be ASIL B for functional safety level. In some examples, the vehicle 1500 may include multiple LiDAR sensors 1564 (e.g., two, four, six, etc.) that may use Ethernet (e.g., to provide data to a Gigabit Ethernet switch).

[0173] In some examples, the LiDAR sensor 1564 may be able to provide a list of objects and their distances for a 360-degree field of view. Commercially available LiDAR sensors 1564 may have, for example, an advertised range of approximately 100m, an accuracy of 2cm-3cm, and support for 100Mbps Ethernet connections. In some examples, one or more non-protruding LiDAR sensors 1564 may be used. In such examples, the LiDAR sensor 1564 may be implemented as a small device that can be embedded in the front, back, side, and / or corner of the vehicle 1500. In such an example, the LiDAR sensor 1564 may provide a field of view of up to 120 degrees horizontally and 35 degrees vertically, with a range of 200m, even for low reflectivity objects. The front-mounted LiDAR sensor 1564 may be configured for a horizontal field of view between 45 and 135 degrees.

[0174] In some examples, LiDAR technologies such as 3D flash LiDAR may also be used. 3D flash LiDAR uses a flash of laser as an emission source to illuminate the vehicle's surroundings up to about 200m. The flash LiDAR unit includes a receiver that records the laser pulse transmission time and the reflected light on each pixel, which in turn corresponds to the range from the vehicle to the object. Flash LiDAR can allow highly accurate and distortion-free images of the surrounding environment to be generated with each laser flash. In some examples, four flash LiDAR sensors can be deployed, one on each side of the vehicle 1500. Available 3D flash LiDAR systems include solid-state 3D staring array LiDAR cameras (e.g., non-scanning LiDAR devices) with no moving parts other than fans. The flash LiDAR device can use 5 nanosecond Class I (eye-safe) laser pulses per frame, and can capture the reflected laser light in the form of a 3D range point cloud and co-registered intensity data. By using flash LiDAR, and because flash LiDAR is a solid-state device with no moving parts, the LiDAR sensor 1564 may be less susceptible to motion blur, vibration, and / or shock.

[0175] The vehicle may further include an IMU sensor 1566. In some examples, the IMU sensor 1566 may be located at the center of the rear axle of the vehicle 1500. The IMU sensor 1566 may include, for example and without limitation, an accelerometer, a magnetometer, a gyroscope, a magnetic compass, and / or other sensor types. In some examples, for example, in a six-axis application, the IMU sensor 1566 may include an accelerometer and a gyroscope, and in a nine-axis application, the IMU sensor 1566 may include an accelerometer, a gyroscope, and a magnetometer.

[0176] In some embodiments, the IMU sensor 1566 can be implemented as a miniature high-performance GPS-aided inertial navigation system (GPS / INS) that combines a micro-electromechanical system (MEMS) inertial sensor, a high-sensitivity GPS receiver, and an advanced Kalman filter algorithm to provide estimates of position, velocity, and attitude. Thus, in some examples, the IMU sensor 1566 can enable the vehicle 1500 to estimate heading by directly observing and correlating velocity changes from the GPS to the IMU sensor 1566 without the need for input from a magnetic sensor. In some examples, the IMU sensor 1566 and the GNSS sensor 1558 can be combined into a single integrated unit.

[0177] The vehicle may include microphones 1596 positioned in and / or around the vehicle 1500. The microphones 1596 may be used for, among other things, emergency vehicle detection and identification.

[0178] The vehicle may further include any number of camera types, including stereo cameras 1568, wide angle cameras 1570, infrared cameras 1572, surround cameras 1574, long-range and / or mid-range cameras 1598, and / or other camera types. These cameras may be used to capture image data around the entire periphery of the vehicle 1500. The type of camera used depends on the embodiment and the requirements of the vehicle 1500, and any combination of camera types may be used to provide the necessary coverage around the vehicle 1500. In addition, the number of cameras may vary depending on the embodiment. For example, the vehicle may include six cameras, seven cameras, ten cameras, twelve cameras, and / or another number of cameras. As an example and not limitation, the cameras may support Gigabit Multimedia Serial Link (GMSL) and / or Gigabit Ethernet. Each of the cameras described herein may include a 10GbE compliant wireless communications interface (NFC) interface, a 10GbE compliant wireless communications interface, and / or a 10GbE compliant wireless communications interface. Fig.15A and Fig. 15B Described in more detail.

[0179] The vehicle 1500 may further include a vibration sensor 1542. The vibration sensor 1542 may measure vibrations of components of the vehicle, such as an axle. For example, changes in vibration may indicate changes in the road surface. In another example, when two or more vibration sensors 1542 are used, the difference between the vibrations may be used to determine friction or slip of the road surface (e.g., when there is a vibration difference between a powered drive shaft and a free-spinning shaft).

[0180] The vehicle 1500 may include an ADAS system 1538. In some examples, the ADAS system 1538 may include a SoC. The ADAS system 1538 may include autonomous / adaptive / automatic cruise control (ACC), cooperative adaptive cruise control (CACC), forward collision warning (FCW), automatic emergency braking (AEB), lane departure warning (LDW), lane keeping assist (LKA), blind spot warning (BSW), rear cross traffic warning (RCTW), collision warning system (CWS), lane centering (LC), and / or other features and functions.

[0181] The ACC system may use a RADAR sensor 1560, a LiDAR sensor 1564, and / or a camera. The ACC system may include a longitudinal ACC and / or a lateral ACC. The longitudinal ACC monitors and controls the distance to the vehicle immediately in front of the vehicle 1500, and automatically adjusts the vehicle speed to maintain a safe distance from the vehicle in front. The lateral ACC performs distance keeping and suggests that the vehicle 1500 change lanes when necessary. The lateral ACC is related to other ADAS applications such as LCA and CWS.

[0182] CACC uses information from other vehicles, which can be received from other vehicles indirectly via a wireless link or through a network connection (e.g., through the Internet) via a network interface 1524 and / or a wireless antenna 1526. A direct link can be provided by a vehicle-to-vehicle (V2V) communication link, while an indirect link can be an infrastructure-to-vehicle (I2V) communication link. Typically, the V2V communication concept provides information about the vehicle immediately ahead (e.g., the vehicle immediately ahead of the vehicle 1500 and in the same lane as it), while the I2V communication concept provides information about traffic farther ahead. The CACC system can include either or both of the I2V and V2V information sources. Given information about the vehicle ahead of the vehicle 1500, CACC can be more reliable, and it is possible to improve the smoothness of traffic flow and reduce road congestion.

[0183] The FCW system is designed to alert the driver to hazards so that the driver can take corrective action. The FCW system uses a front camera and / or RADAR sensor 1560 coupled to a dedicated processor, DSP, FPGA and / or ASIC, which is electrically coupled to driver feedback such as a display, speaker and / or vibrating components. The FCW system can provide warnings in the form of, for example, sound, visual warnings, vibrations and / or rapid brake pulses.

[0184] The AEB system detects an impending front collision with another vehicle or other object and can automatically apply the brakes if the driver does not take corrective action within specified time or distance parameters. The AEB system can use a front camera and / or RADAR sensor 1560 coupled to a dedicated processor, DSP, FPGA and / or ASIC. When the AEB system detects a hazard, it typically first alerts the driver to take corrective action to avoid the collision, and if the driver does not take corrective action, the AEB system can automatically apply the brakes in an effort to prevent or at least mitigate the effects of the predicted collision. The AEB system may include technologies such as dynamic brake support and / or collision approach braking.

[0185] The LDW system provides visual, audible, and / or tactile warnings, such as steering wheel or seat vibrations, to alert the driver when the vehicle 1500 crosses a lane marking. When the driver indicates an intention to leave the lane, by activating a turn signal, the LDW system is not activated. The LDW system may use a front-facing camera coupled to a dedicated processor, DSP, FPGA, and / or ASIC that is electrically coupled to driver feedback such as a display, speaker, and / or vibrating component.

[0186] The LKA system is a variation of the LDW system. If the vehicle 1500 begins to leave the lane, the LKA system provides steering input or braking to correct the vehicle 1500.

[0187] The BSW system detects and warns the driver of vehicles in the car's blind spot. The BSW system can provide visual, auditory and / or tactile alerts to indicate that it is unsafe to merge or change lanes. The system can provide additional warnings when the driver uses a turn signal. The BSW system can use a rear-facing camera and / or RADAR sensor 1560 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 components.

[0188] The RCTW system may provide visual, audible, and / or tactile notifications when an object is detected outside the range of the rear camera while the vehicle 1500 is in reverse. Some RCTW systems include AEB to ensure that the vehicle brakes are applied to avoid a crash. The RCTW system may use one or more rear RADAR sensors 1560 coupled to a dedicated processor, DSP, FPGA, and / or ASIC that is electrically coupled to driver feedback such as a display, speaker, and / or vibrating component.

[0189] Conventional ADAS systems may be prone to false positive results, which may annoy and distract the driver, but are typically not catastrophic because the ADAS system alerts the driver and allows the driver to decide whether the safety condition actually exists and take action accordingly. However, in the autonomous vehicle 1500, in the case of conflicting results, the vehicle 1500 itself must decide whether to pay attention to the results from the main computer or the auxiliary computer (e.g., the first controller 1536 or the second controller 1536). For example, in some embodiments, the ADAS system 1538 can be a backup and / or auxiliary computer for providing perception information to the backup computer rationality module. The backup computer rationality monitor can run redundant and diverse software on hardware components to detect faults in perception and dynamic driving tasks. The output from the ADAS system 1538 can be provided to the supervisory MCU. If the outputs from the main computer and the auxiliary computer conflict, the supervisory MCU must determine how to coordinate the conflict to ensure safe operation.

[0190] In some examples, the master computer can be configured to provide a confidence score to the supervisory MCU, indicating the master computer's confidence in the selected result. If the confidence score exceeds a threshold, the supervisory MCU can follow the direction of the master computer, regardless of whether the auxiliary computer provides conflicting or inconsistent results. In the case where the confidence score does not meet the threshold and the master computer and the auxiliary computer indicate different results (e.g., conflicts), the supervisory MCU can arbitrate between these computers to determine the appropriate result.

[0191] The supervisory MCU may be configured to run a neural network that is trained and configured to determine conditions under which the auxiliary computer provides a false alarm based on outputs from the primary computer and the auxiliary computer. Thus, the neural network in the supervisory MCU may learn when the output of the auxiliary computer may be trusted and when it may not. For example, when the auxiliary computer is a RADAR-based FCW system, the neural network in the supervisory MCU may learn when the FCW system is identifying a metal object that is not actually dangerous, such as a drainage grate or manhole cover that triggers an alarm. Similarly, when the auxiliary computer is a camera-based LDW system, the neural network in the supervisory MCU may learn to ignore the LDW when a cyclist or pedestrian is present and lane departure is actually the safest strategy. In an embodiment that includes a neural network running on the supervisory MCU, the supervisory MCU may include at least one of a DLA or GPU suitable for running the neural network with associated memory. In a preferred embodiment, the supervisory MCU may include and / or be included as a component of SoC 1504.

[0192] In other examples, the ADAS system 1538 may include an auxiliary computer that uses traditional computer vision rules to perform ADAS functions. In this way, the auxiliary computer can use classic computer vision rules (if-then), and the presence of a neural network in the supervisory MCU can improve reliability, safety, and performance. For example, diverse implementations and intentional non-identity make the entire system more fault-tolerant, especially for failures caused by software (or software-hardware interface) functions. For example, if there is a software vulnerability or error in the software running on the main computer and the non-identical software code running on the auxiliary computer provides the same overall result, the supervisory MCU can be more confident that the overall result is correct and that the vulnerability in the software or hardware on the main computer does not cause a substantial error.

[0193] In some examples, the output of the ADAS system 1538 can be fed to the perception block of the main computer and / or the dynamic driving task block of the main computer. For example, if the ADAS system 1538 indicates a forward collision warning due to an object immediately ahead, the perception block can use this information in identifying the object. In other examples, the auxiliary computer can have its own neural network that is trained and thus reduces the risk of false positives as described herein.

[0194] The vehicle 1500 may further include an infotainment SoC 1530 (e.g., an in-vehicle infotainment system (IVI)). Although illustrated and described as an SoC, the infotainment system may not be an SoC and may include two or more discrete components. The infotainment SoC 1530 may include a combination of hardware and software that may be used to provide audio (e.g., music, personal digital assistant, navigation instructions, news, radio, etc.), video (e.g., TV, movies, streaming, etc.), phone (e.g., hands-free calling), network connectivity (e.g., LTE, Wi-Fi, etc.), and / or information services (e.g., navigation system, rear parking assistance, radio data system, vehicle-related information such as fuel level, total distance covered, brake fuel level, oil level, door open / closed, air filter information, etc.) to the vehicle 1500. For example, the infotainment SoC 1530 may include a radio, a disc player, a navigation system, a video player, USB and Bluetooth connectivity, an onboard computer, onboard entertainment, Wi-Fi, steering wheel audio controls, hands-free voice controls, a head-up display (HUD), an HMI display 1534, 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 1530 may further be used to provide information (e.g., visual and / or auditory) to a user of the vehicle, such as information from an ADAS system 1538, 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.

[0195] The infotainment SoC 1530 may include GPU functionality. The infotainment SoC 1530 may communicate with other devices, systems, and / or components of the vehicle 1500 via a bus 1502 (e.g., a CAN bus, Ethernet, etc.). In some examples, the infotainment SoC 1530 may be coupled to a supervisory MCU so that in the event of a failure of a master controller 1536 (e.g., a primary and / or backup computer of the vehicle 1500), the GPU of the infotainment system may perform some self-driving functions. In such an example, the infotainment SoC 1530 may place the vehicle 1500 in a driver-safe parking mode as described herein.

[0196] The vehicle 1500 may further include an instrument cluster 1532 (e.g., a digital instrument panel, an electronic instrument cluster, a digital instrument panel, etc.). The instrument cluster 1532 may include a controller and / or a supercomputer (e.g., a discrete controller or a supercomputer). The instrument cluster 1532 may include a set of instruments, such as a speedometer, a fuel level, an oil pressure, a tachometer, an odometer, a turn indicator, a shift position indicator, a seat belt warning light, a parking brake warning light, an engine fault light, an airbag (SRS) system information, a lighting control, a safety system control, navigation information, and the like. In some examples, information may be displayed and / or shared between the infotainment SoC 1530 and the instrument cluster 1532. In other words, the instrument cluster 1532 may be included as part of the infotainment SoC 1530, or vice versa.

[0197] Fig.15D For a cloud-based server and Fig.15A 1576. System 1576 may include server 1578, network 1590, and vehicles including vehicle 1500. Server 1578 may include multiple GPUs 1584(A)-1584(H) (collectively referred to herein as GPU 1584), PCIe switches 1582(A)-1582(H) (collectively referred to herein as PCIe switches 1582), and / or CPUs 1580(A)-1580(B) (collectively referred to herein as CPU 1580). GPU 1584, CPU 1580, and PCIe switches may be interconnected with a high-speed interconnect and / or PCIe connection 1586 such as, for example and without limitation, NVLink interface 1588 developed by NVIDIA. In some examples, GPU 1584 is connected via NVLink and / or NVSwitch SoC, and GPU 1584 and PCIe switch 1582 are connected via PCIe interconnect. Although eight GPUs 1584, two CPUs 1580, and two PCIe switches are illustrated, this is not intended to be limiting. Depending on the embodiment, each of the servers 1578 may include any number of GPUs 1584, CPUs 1580, and / or PCIe switches. For example, each of the servers 1578 may include eight, sixteen, thirty-two, and / or more GPUs 1584.

[0198] Server 1578 may receive image data over network 1590 and from a vehicle, the image data representing images showing unexpected or changed road conditions, such as recently begun road work. Server 1578 may transmit neural network 1592, updated neural network 1592, and / or map information 1594, including information about traffic and road conditions, over network 1590 and to the vehicle. Updates to map information 1594 may include updates to HD map 1522, such as information about construction sites, potholes, curves, flooding, or other obstacles. In some examples, neural network 1592, updated neural network 1592, and / or map information 1594 may have been generated from new training and / or data received from any number of vehicles in the environment and / or based on experience from training performed at a data center (e.g., using server 1578 and / or other servers).

[0199] Server 1578 can be used to train a machine learning model (e.g., a neural network) based on training data. The training data can be generated by the vehicle, and / or can be generated in a simulation (e.g., using a game engine). In some examples, the training data is labeled (e.g., when the neural network benefits from supervised learning) and / or undergoes other preprocessing, while in other examples, the training data is not labeled and / or preprocessed (e.g., when the neural network does not require supervised learning). Training can be performed according to any one or more categories of machine learning techniques, including but not limited to: categories such as supervised training, semi-supervised training, unsupervised training, self-learning, reinforcement learning, joint learning, transfer learning, feature learning (including main components and cluster analysis), multilinear subspace learning, manifold learning, representation learning (including alternate 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 via network 1590), and / or the machine learning model can be used by server 1578 to remotely monitor the vehicle.

[0200] In some examples, server 1578 can receive data from the vehicle and apply the data to the latest real-time neural network for real-time intelligent reasoning. Server 1578 may include a deep learning supercomputer and / or a dedicated AI computer powered by GPU 1584, such as DGX and DGX Station machines developed by NVIDIA. However, in some examples, server 1578 may include a deep learning infrastructure of a data center powered only by CPUs.

[0201] The deep learning infrastructure of server 1578 may be capable of rapid real-time inference, and may use this capability to assess and verify the health of the processors, software, and / or associated hardware in vehicle 1500. For example, the deep learning infrastructure may receive periodic updates from vehicle 1500, such as a sequence of images and / or objects located in the sequence of images that vehicle 1500 has located (e.g., via computer vision and / or other machine learning object classification techniques). The deep learning infrastructure may run its own neural network to identify objects and compare them to the objects identified by vehicle 1500, and if the results do not match and the infrastructure concludes that the AI ​​in vehicle 1500 has failed, then server 1578 may transmit a signal to vehicle 1500 instructing the fail-safe computer of vehicle 1500 to take control, notify passengers, and complete a safe parking maneuver.

[0202] For reasoning, server 1578 may include GPU 1584 and one or more programmable reasoning accelerators (e.g., NVIDIA's TensorRT). The combination of GPU-powered servers and reasoning acceleration can make real-time responses possible. In other examples, such as when performance is not so important, CPU, FPGA, and other processor-powered servers can be used for reasoning.

[0203] Example computing device

[0204] Fig.16 1 is a block diagram of an example computing device 1600 suitable for implementing some embodiments of the present disclosure. The computing device 1600 may include an interconnect system 1602 that directly or indirectly couples the following devices: a memory 1604, one or more central processing units (CPUs) 1606, one or more graphics processing units (GPUs) 1608, a communication interface 1610, an input / output (I / O) port 1612, an input / output component 1614, a power supply 1616, one or more presentation components 1618 (e.g., a display), and one or more logic units 1620. In at least one embodiment, the computing device 1600 may include one or more virtual machines (VMs), and / or any of its components may include virtual components (e.g., virtual hardware components). For non-limiting examples, the one or more GPUs 1608 may include one or more vGPUs, the one or more CPUs 1606 may include one or more vCPUs, and / or the one or more logic units 1620 may include one or more virtual logic units. Thus, computing device 1600 may include discrete components (eg, a complete GPU dedicated to computing device 1600 ), virtual components (eg, a portion of a GPU dedicated to computing device 1600 ), or a combination thereof.

[0205] although Fig.16The various blocks of are shown as being connected via an interconnect system 1602 having wires, but this is not intended to be limiting and is merely for clarity. For example, in some embodiments, a presentation component 1618 such as a display device may be considered an I / O component 1614 (e.g., if the display is a touch screen). As another example, CPU 1606 and / or GPU 1608 may include memory (e.g., memory 1604 may represent a storage device in addition to the memory of GPU 1608, CPU 1606, and / or other components). In other words, Fig.16 The computing devices described herein are illustrative only. No distinction is made between categories such as "workstations," "servers," "laptops," "desktops," "tablets," "client devices," "mobile devices," "handheld devices," "game consoles," "electronic control units (ECUs)," "virtual reality systems," and / or other device or system types, as all are considered Fig.16 within the range of computing devices.

[0206] The interconnection system 1602 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 1602 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 standard association (VESA) bus, a peripheral component interconnect (PCI) bus, a peripheral component interconnect express (PCIe) bus, and / or another type of bus or link. In some embodiments, there is a direct connection between components. For example, CPU 1606 may be directly connected to memory 1604. In addition, CPU 1606 may be directly connected to GPU 1608. In the case where there is a direct or point-to-point connection between components, the interconnection system 1602 may include a PCIe link to perform the connection. In these examples, the PCI bus does not need to be included in the computing device 1600.

[0207] Memory 1604 may include any of a variety of computer-readable media. Computer-readable media may be any available media that can be accessed by computing device 1600. Computer-readable media may include volatile and nonvolatile media and removable and non-removable media. By way of example and not limitation, computer-readable media may include computer storage media and communication media.

[0208] Computer storage media may include volatile and nonvolatile media and / or removable and non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, program modules, and / or other data types. For example, memory 1604 may store computer-readable instructions (e.g., representing programs and / or program elements, such as an operating system). Computer storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other storage technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by computing device 1600. As used herein, computer storage media does not include the signals themselves.

[0209] Computer storage media may include computer readable instructions, data structures, program modules, and / or other data types in a modulated data signal such as a carrier wave or other transmission mechanism, and include any information delivery media. The term "modulated data signal" may refer to a signal that has one or more of its characteristics set or changed in such a way that information is encoded into the signal. By way of example and not limitation, computer storage media may include wired media such as a wired network or a direct wired connection, and wireless media such as sound, RF, infrared, and other wireless media. Any combination of the above should also be included within the scope of computer readable media.

[0210] The CPU 1606 may be configured to execute at least some of the computer-readable instructions so as to control one or more components of the computing device 1600 to perform one or more of the methods and / or processes described herein. Each of the CPUs 1606 may include one or more cores (e.g., one, two, four, eight, twenty-eight, seventy-two, etc.) capable of simultaneously processing a large number of software threads. The CPU 1606 may include any type of processor, and may include different types of processors, depending on the type of computing device 1600 implemented (e.g., a processor with fewer cores for mobile devices and a processor with more cores for servers). For example, depending on the type of computing device 1600, 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). The computing device 1600 may also include one or more CPUs 1606 in addition to one or more microprocessors or supplementary coprocessors such as math coprocessors.

[0211] In addition to or as an alternative to CPU 1606, one or more GPUs 1608 may be configured to execute at least some of the computer-readable instructions to control one or more components of computing device 1600 to perform one or more of the methods and / or processes described herein. One or more GPUs 1608 may be integrated GPUs (e.g., with one or more CPUs 1606) and / or one or more GPUs 1608 may be discrete GPUs. In an embodiment, one or more GPUs 1608 may be coprocessors of one or more CPUs 1606. Computing device 1600 may use GPU 1608 to render graphics (e.g., 3D graphics) or perform general purpose computing. For example, one or more GPUs 1608 may be used for general purpose computing (GPGPU) on GPUs. One or more GPUs 1608 may include hundreds of thousands of cores capable of processing hundreds of thousands of software threads simultaneously. GPU 1608 may generate pixel data for outputting an image in response to a rendering command (e.g., a rendering command from CPU 1606 received via a host interface). GPU 1608 may include graphics memory, such as display memory, for storing pixel data or any other suitable data, such as GPGPU data. Display memory may be included as part of memory 1604. One or more GPUs 1608 may include two or more GPUs operating in parallel (e.g., via a link). The link may connect the GPUs directly (e.g., using NVLINK) or may connect the GPUs through a switch (e.g., using NVSwitch). When combined together, each GPU 1608 may generate pixel data or GPGPU data for different parts of the output or different outputs (e.g., a first GPU for a first image and a second GPU for a second image). Each GPU may include its own memory or may share memory with other GPUs.

[0212] In addition to or as an alternative to the CPU 1606 and / or GPU 1608, the logic unit 1620 may be configured to execute at least some of the computer-readable instructions to control one or more components of the computing device 1600 to perform one or more of the methods and / or processes described herein. In an embodiment, the CPU 1606, the GPU 1608, and / or the logic unit 1620 may perform any combination of methods, processes, and / or portions thereof separately or jointly. The one or more logic units 1620 may be a part of and / or integrated into one or more of the CPU 1606 and / or GPU 1608, and / or the one or more logic units 1620 may be a discrete component or otherwise external to the CPU 1606 and / or GPU 1608. In an embodiment, the one or more logic units 1620 may be a coprocessor of the one or more CPUs 1606 and / or the one or more GPUs 1608.

[0213] Examples of logic unit 1620 include one or more processing cores and / or components thereof, such as a data processing unit (DPU), a tensor core (TC), a tensor processing unit (TPU), a pixel vision core (PVC), a vision processing unit (VPU), a graphics processing cluster (GPC), a texture processing cluster (TPC), a streaming multiprocessor (SM), a tree traversal unit (TTU), an artificial intelligence accelerator (AIA), a deep learning accelerator (DLA), an arithmetic logic unit (ALU), an application specific integrated circuit (ASIC), a floating point unit (FPU), an input / output (I / O) element, a peripheral component interconnect (PCI) or a peripheral component interconnect express (PCIe) element, etc.

[0214] The communication interface 1610 may include one or more receivers, transmitters, and / or transceivers that enable the computing device 1600 to communicate with other computing devices via an electronic communication network, including wired and / or wireless communications. The communication interface 1610 may include components and functionality that enable communication over any of a number of different networks, such as a wireless network (e.g., Wi-Fi, Z-wave, Bluetooth, Bluetooth LE, ZigBee, etc.), a wired network (e.g., communicating over Ethernet or InfiniBand), a low power wide area network (e.g., LoRaWAN, SigFox, etc.), and / or the Internet. In one or more embodiments, the logic unit 1620 and / or the communication interface 1610 may include one or more data processing units (DPUs) to transmit data received over a network and / or over the interconnect system 1602 directly to one or more GPUs 1608 (e.g., their memory).

[0215] I / O ports 1612 can enable computing device 1600 to be logically coupled to other devices including I / O components 1614, presentation components 1618, and / or other components, some of which can be built into (e.g., integrated into) computing device 1600. Illustrative I / O components 1614 include microphones, mice, keyboards, joysticks, game pads, game controllers, satellite dishes, scanners, printers, wireless devices, and the like. I / O components 1614 can provide a natural user interface (NUI) that processes air gestures, voice, or other physiological inputs generated by a user. In some instances, the input can be transmitted to an appropriate network element for further processing. The NUI can implement any combination of voice recognition, stylus recognition, facial recognition, biometric recognition, gesture recognition on and adjacent to the screen, air gestures, head and eye tracking, and touch recognition associated with the display of computing device 1600 (as described in more detail below). Computing device 1600 may include a depth camera such as a stereo camera system, an infrared camera system, an RGB camera system, touch screen technology, and combinations of these for gesture detection and recognition. Additionally, computing device 1600 may include an accelerometer or gyroscope to enable motion detection (e.g., as part of an inertial measurement unit (IMU)). In some examples, the output of the accelerometer or gyroscope may be used by computing device 1600 to render immersive augmented reality or virtual reality.

[0216] The power supply 1616 may include a hardwired power supply, a battery power supply, or a combination thereof. The power supply 1616 may provide power to the computing device 1600 to enable the components of the computing device 1600 to operate.

[0217] The presentation component 1618 may include a display (e.g., a monitor, a touch screen, a television screen, a head-up display (HUD), other display types, or a combination thereof), a speaker, and / or other presentation components. The presentation component 1618 may receive data from other components (e.g., GPU 1608, CPU 1606, DPU, etc.) and output the data (e.g., as an image, video, sound, etc.).

[0218] Sample Data Center

[0219] Fig.17 An example data center 1700 that can be used in at least one embodiment of the present disclosure is shown. The data center 1700 can include a data center infrastructure layer 1710, a framework layer 1720, a software layer 1730, and / or an application layer 1740.

[0220] like Fig.17As shown, the data center infrastructure layer 1710 may include a resource coordinator 1712, grouped computing resources 1714, and node computing resources ("node CRs") 1716(1)-1716(N), where "N" represents any complete positive integer. In at least one embodiment, the node CRs 1716(1)-1716(N) may include, but are not limited to, any number of central processing units ("CPUs") or other processors (including DPUs, accelerators, field programmable gate arrays (FPGAs), graphics processors or graphics processing units (GPUs), etc.), memory devices (e.g., dynamic read-only memories), storage devices (e.g., solid-state or disk drives), network input / output ("NW I / O") devices, network switches, virtual machines ("VMs"), power modules and / or cooling modules, etc. In some embodiments, one or more of the node CRs 1716(1)-1716(N) may correspond to a server having one or more of the above-mentioned computing resources. Furthermore, in some embodiments, node CRs 1716(1)-17161(N) may include one or more virtual components, such as vGPU, vCPU, etc., and / or one or more of node CRs 1916(1)-1916(N) may correspond to a virtual machine (VM).

[0221] In at least one embodiment, the computing resources 1714 of the grouping may include a separate grouping of node CR1716 housed in one or more racks (not shown), or many racks housed in a data center at different geographical locations (also not shown). The separate grouping of node CR1716 in the computing resources 1714 of the grouping may include grouping computing, network, memory or storage resources that may be configured or allocated to support one or more workloads. In at least one embodiment, several node CR1716 including CPU, GPU, DPU 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.

[0222] Resource coordinator 1712 may configure or otherwise control one or more nodes CR 1716(1)-1716(N) and / or grouped computing resources 1714. In at least one embodiment, resource coordinator 1712 may include a software design infrastructure ("SDI") management entity for data center 1700. Resource coordinator 1712 may include hardware, software, or some combination thereof.

[0223] In at least one embodiment, Fig.17As shown, the framework layer 1720 may include a job scheduler 1733, a configuration manager 1734, a resource manager 1736, and / or a distributed file system 1738. The framework layer 1720 may include a framework that supports software 1732 of the software layer 1730 and / or one or more applications 1742 of the application layer 1740. The software 1732 or the application 1742 may include network-based service software or applications, such as those provided by Amazon Web Services, Google Cloud, and Microsoft Azure. The framework layer 1720 may be, but is not limited to, a free and open source software network application framework (such as Apache Spark) that can utilize the distributed file system 1738 for large-scale data processing (e.g., "big data"). TM (hereinafter referred to as "Spark")). In at least one embodiment, the job scheduler 1733 may include a Spark driver to facilitate scheduling workloads supported by different layers of the data center 1700. The configuration manager 1734 may be able to configure different layers, such as the software layer 1730 and the framework layer 1720 (which includes Spark and a distributed file system 1738 for supporting large-scale data processing). The resource manager 1736 may be able to manage clustered or grouped computing resources mapped to the distributed file system 1778 and the job scheduler 1733 or allocated to support the distributed file system 1738 and the job scheduler 1733. In at least one embodiment, the clustered or grouped computing resources may include grouped computing resources 1714 at the data center infrastructure layer 1710. The resource manager 1736 may coordinate with the resource coordinator 1712 to manage these mapped or allocated computing resources.

[0224] In at least one embodiment, the software 1732 included in software layer 1730 may include software used by at least a portion of node CRs 1716(1)-1716(N), grouped computing resources 1714, and / or distributed file system 1738 of framework layer 1720. One or more types of software may include, but are not limited to, Internet web page search software, email virus scanning software, database software, and streaming video content software.

[0225] In at least one embodiment, the applications 1742 included in the application layer 1740 may include one or more types of applications used by at least a portion of the node CRs 1716(1)-1716(N), the grouped computing resources 1714, and / or the distributed file system 1738 of the framework layer 1720. The one or more types of applications may include, but are not limited to, any number of genomic applications, cognitive computing, and machine learning applications, including training or inference software, machine learning framework software (e.g., PyTorch, TensorFlow, Caffe, etc.), and / or other machine learning applications used in conjunction with one or more embodiments.

[0226] In at least one embodiment, any of configuration manager 1734, resource manager 1736, and resource coordinator 1712 can implement any number and type of self-modification actions based on any amount and type of data acquired in any technically feasible manner. The self-modification actions can save a data center operator of data center 1700 from making potentially poor configuration decisions and potentially avoiding underutilized and / or poorly performing portions of the data center.

[0227] According to one or more embodiments described herein, data center 1700 may include tools, services, software, or other resources to train one or more machine learning models or to predict or infer information using one or more machine learning models. For example, the machine learning model(s) may be trained by calculating weight parameters according to a neural network architecture using the software and / or computing resources described above with respect to data center 1700. In at least one embodiment, a trained or deployed machine learning model corresponding to one or more neural networks may be used to infer or predict information using the resources described above with respect to data center 1700 using weight parameters calculated using one or more training techniques, such as, but not limited to, those described herein.

[0228] In at least one embodiment, the data center 1700 may use a CPU, an application-specific integrated circuit (ASIC), a GPU, an FPGA, and / or other hardware (or virtual computing resources corresponding thereto) to perform training and / or reasoning using the above resources. In addition, one or more software and / or hardware resources described above may be configured to allow a user to train or perform reasoning services on information, such as image recognition, speech recognition, or other artificial intelligence services.

[0229] Example network environment

[0230] A network environment suitable for implementing embodiments of the present disclosure may include one or more client devices, servers, network attached storage (NAS), other backend devices, and / or other device types. The client devices, servers, and / or other device types (e.g., each device) may be configured to communicate with the client device, server, or other device types. Fig.16 The backend device 1700 may be implemented on one or more instances of the computing device 1600 of the backend device 1600 - for example, each device may include similar components, features and / or functions of the computing device 1600. In addition, in the case of implementing a backend device (e.g., a server, NAS, etc.), the backend device may be included as part of the data center 1700, and examples of the data center 1700 are described herein with reference to FIG. Fig.17 Describe in more detail.

[0231] The components of the network environment can communicate with each other via a network, which can be wired, wireless, or both. The network can include multiple networks or one of multiple networks. For example, the network can include one or more wide area networks (WANs), one or more local area networks (LANs), one or more public networks (such as the Internet and / or 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 (and other components) can provide wireless connections.

[0232] 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 for the server may be implemented on any number of client devices.

[0233] In at least one embodiment, the network environment may include one or more cloud-based network environments, distributed computing environments, combinations thereof, and the like. A cloud-based network environment may include a framework layer, a job scheduler, a resource manager, and a distributed file system implemented on one or more servers, and the server may include one or more core network servers and / or edge servers. The framework layer may include a framework for software supporting the software layer and / or one or more applications of the application layer. The software or application may include network-based service software or applications, respectively. In an embodiment, one or more client devices may use network-based service software or applications (e.g., by accessing the service software and / or application via one or more application programming interfaces (APIs)). The framework layer may be, but is not limited to, a free and open source software network application framework that can use a distributed file system for large-scale data processing (e.g., "big data").

[0234] 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 different functions can be distributed across multiple locations from a central or core server (e.g., one or more data centers that can be distributed across a state, region, country, global, etc.). If the connection to the user (e.g., client device) is relatively close to an edge server, the core server can assign at least a portion of the functionality to the edge server. A cloud-based network environment can be private (e.g., limited to a single organization), public (e.g., available to many organizations), and / or a combination thereof (e.g., a hybrid cloud environment).

[0235] The client device(s) may include the Fig.16 At least some of the components, features, and functionality of the described example computing device(s) 1600. By way of example and not limitation, a client device may be implemented as a personal computer (PC), a laptop computer, a mobile device, a smart phone, a tablet computer, a smart watch, a wearable computer, a personal digital assistant (PDA), an MP3 player, a virtual reality headset, a global positioning system (GPS) or device, a video player, a camera, a surveillance device or system, a vehicle, a boat, a spacecraft, 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.

[0236] The present disclosure may be described in the general context of machine-usable instructions or computer codes executed by a computer or other machine such as a personal digital assistant or other handheld device, including computer-executable instructions such as program modules. Typically, program modules including routines, programs, objects, components, data structures, etc. refer to codes that perform specific tasks or implement specific abstract data types. The present disclosure may be practiced in a variety of system configurations, including handheld devices, consumer electronics, general-purpose computers, more specialized computing devices, etc. The present disclosure may also be practiced in a distributed computing environment in which tasks are performed by remote processing devices linked through a communication network.

[0237] As used herein, the description of "and / or" about two or more elements should be interpreted as referring to only one element or a combination of elements. For example, "element A, element B and / or element C" may include only element A, only element B, only element C, element A and element B, element A and element C, element B and element C, or elements A, B and C. In addition, "at least one of element A or element B" may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B. Further, "at least one of element A and element B" may include at least one of element A, at least one of element B, or at least one of element A and at least one of element B.

[0238] The subject matter of the present disclosure is described in detail herein to meet statutory requirements. However, the description itself is not intended to limit the scope of the present disclosure. On the contrary, the inventors have contemplated that the claimed subject matter may also be embodied in other ways to include steps different from the steps described herein in conjunction with other current or future technologies or combinations of similar steps. Moreover, although the terms "step" and / or "block" may be used herein to imply different elements of the method employed, these terms should not be interpreted as implying any particular order among or between the various steps disclosed herein, unless the order of the steps is explicitly described.

Claims

1. A method comprising: determining a first posture associated with the machine based at least on motion data obtained using one or more motion sensors of the machine; determining one or more differences between the first pose and a second pose associated with the machine, the second pose being determined using sensor data obtained using one or more sensors of the machine; determining one or more correspondences between one or more first features represented by the sensor data and one or more second features represented by the map data; determining an error associated with the second pose based at least on the one or more differences and the one or more correspondences; and Based at least on the error, the machine is caused to perform one or more operations.

2. The method of claim 1, further comprising: An uncertainty associated with the error is determined based at least on at least one of the one or more differences or the one or more correspondences.

3. The method of claim 1, further comprising: comparing the one or more first features represented by the sensor data with the one or more second features represented by the map data; determining one or more costs based at least on the comparing; as well as The second posture associated with the machine is determined based at least on the one or more costs.

4. The method of claim 1 , wherein determining the first posture associated with the machine comprises: determining a third posture associated with the machine; determining a motion of the machine based at least on the motion data, the motion of the machine comprising at least one of: a direction of travel of the machine, a distance traveled by the machine, or a change in a yaw angle associated with the machine; as well as The first posture associated with the machine is determined based on at least the third posture associated with the machine and the motion of the machine.

5. The method of claim 1 , wherein determining the error associated with the second pose comprises: determining a first error based at least on the one or more differences; determining a second error based at least on the one or more corresponding relationships; as well as The error is determined based on at least the first error and the second error.

6. The method of claim 1 , wherein determining the one or more corresponding relationships between the one or more first features represented by the sensor data and the one or more second features represented by the map data comprises: determining that the sensor data is representative of the one or more first characteristics; as well as A determination is made as to whether the one or more first features represented by the sensor data match the one or more second features represented by the map data.

7. The method of claim 1, further comprising: determining whether the error is less than or equal to an error threshold; as well as Whether to cause the machine to use the second posture when performing the one or more operations is determined based at least on whether the error is less than or equal to the error threshold.

8. The method of claim 1, further comprising: determining a third posture associated with the machine based at least on second motion data obtained using the one or more motion sensors of the machine; determining one or more second differences between the third pose and a fourth pose associated with the machine, the fourth pose determined using second sensor data obtained using the one or more sensors of the machine; determining one or more second correspondences between one or more third features represented by the second sensor data and one or more fourth features represented by the map data; as well as determining a second error associated with the fourth gesture based at least on the one or more second differences and the one or more second correspondences, Wherein determining the error is also based at least on the second error.

9. The method of claim 1, wherein the error comprises one or more of: a first error associated with the x-coordinate direction; a second error associated with the y-coordinate direction; a third error associated with the z-coordinate direction; a fourth error associated with yaw; a fifth error associated with roll; or A sixth error is associated with pitch.

10. A system comprising: One or more processing units, the one or more processing units being configured to: determining one or more differences between a first pose determined using the motion data and a second pose determined using the sensor data; determining one or more correspondences between one or more first features represented by the sensor data and one or more second features represented by the map data; determining an error associated with the second pose based at least on the one or more differences and the one or more correspondences; and Based at least on the error, the machine is caused to perform one or more operations.

11. The system of claim 10, wherein the one or more processing units are further configured to determine an uncertainty associated with the error based at least on at least one of the one or more differences or the one or more correspondences.

12. The system of claim 10, wherein the one or more processing units are further configured to: comparing the one or more first features represented by the sensor data with the one or more second features represented by the map data; determining one or more costs based at least on the comparing; and The second posture associated with the machine is determined based at least on the one or more costs.

13. The system of claim 10, wherein the one or more processing units are further configured to: determining a third posture associated with the machine; determining a motion of the machine based at least on the motion data, the motion of the machine comprising at least one of: a direction of travel of the machine, a distance traveled by the machine, or a change in a yaw angle associated with the machine; and The first posture associated with the machine is determined based on at least the third posture associated with the machine and the motion of the machine.

14. The system of claim 10, wherein determining the error associated with the second pose comprises: determining a first error based at least on the one or more differences; determining a second error based at least on the one or more corresponding relationships; as well as The error is determined based on at least the first error and the second error.

15. The system of claim 10, wherein the one or more processing units are further configured to: determining whether the error is less than or equal to an error threshold; and Whether to cause the machine to use the second posture when performing the one or more operations is determined based at least on whether the error is less than or equal to the error threshold.

16. The system of claim 1, further comprising: determining one or more second differences between a third pose determined using the second motion data and a fourth pose determined using the second sensor data; determining one or more second correspondences between one or more third features represented by the second sensor data and one or more fourth features represented by the map data; as well as determining a second error associated with the fourth gesture based at least on the one or more second differences and the one or more second correspondences, Wherein the error is further determined based at least on the second error.

17. The system of claim 10, wherein the system 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 digital twin operations; A system for performing light transport simulations; A system for performing collaborative content creation of 3D assets; Systems for performing deep learning operations; Systems implemented using edge devices; Systems implemented using robots; A system implemented using a large language model (LLM); A system for performing one or more generative AI operations; Systems for performing conversational AI operations; Systems for generating synthetic data; A system comprising one or more virtual machines VM; A system implemented at least in part in a data center; or A system implemented at least in part using cloud computing resources.

18. A processor, comprising: One or more processing units, the one or more processing units being configured to cause a machine to perform one or more operations based at least on an error associated with a first posture of the machine within an environment, wherein the error is determined based at least on one or more differences between the first posture and a second posture determined using motion data and one or more correspondences between one or more first features represented by sensor data and one or more second features represented by map data.

19. The processor of claim 18, wherein the one or more processing units are further configured to determine an uncertainty associated with the error based at least on at least one of the one or more differences or the one or more correspondences.

20. The processor of claim 18, 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 digital twin operations; A system for performing light transport simulations; A system for performing collaborative content creation of 3D assets; Systems for performing deep learning operations; Systems implemented using edge devices; Systems implemented using robots; A system implemented using a large language model (LLM); A system for performing one or more generative AI operations; Systems for performing conversational AI operations; Systems for generating synthetic data; A system comprising one or more virtual machines VM; A system implemented at least in part in a data center; or A system implemented at least in part using cloud computing resources.

Images