Abstract: In one embodiment, a method by a computing system of a vehicle includes determining an environment of the vehicle. The method includes generating, based on the environment, multiple proposed vehicle actions with associated operational data. The method includes determining a comfort level for each proposed vehicle action by processing the environment and operational data using a model for predicting comfort levels of vehicle actions. The model is trained using records of performed vehicle actions. The record for each performed vehicle action includes environment data, operational data, and a perceived passenger comfort level for the performed vehicle action. The method includes selecting a vehicle action from the proposed vehicle actions based on the determined comfort level. The method includes causing the vehicle to perform the selected vehicle action.

Abstract: In one embodiment, a computing system of a vehicle generates perception data based on sensor data captured by one or more sensors of the vehicle. The perception data includes one or more representations of physical objects in an environment associated with the vehicle. The computing system further determines simulated perception data that includes one or more representations of virtual objects within the environment and generates modified perception data based on the perception data and the simulated perception data. The modified perception data includes at least one of the one or more representations of physical objects and the one or more representations of virtual objects. The computing system further determines a path of travel for the vehicle based on the modified perception data, which includes the one or more representations of the virtual objects.

Abstract: In one embodiment, a method includes, by a computing system of a vehicle, determining a first temperature of a control unit of the vehicle, wherein the control unit and one or more components of the vehicle are interconnected through a thermal network, determining a thermal objective for the control unit based on the first temperature of the control unit, selecting, based on the thermal objective for the control unit, at least a first component from the one or more components, and sending one or more signals to one or more actuators within the thermal network to enable a heat-transfer fluid to flow between (1) a first portion of the thermal network thermally coupled to the control unit and (2) a second portion of the thermal network thermally coupled to the selected first component.

Abstract: In one embodiment, a method includes receiving sensor data of an environment of the vehicle generated by one or more sensors of the vehicle, the sensors comprising a camera, identifying, based on the sensor data, one or more objects in a field of view of the camera and one or more object types that correspond to the one or more objects, determining one or more target histograms that correspond to the object types, generating a processed image based on an image captured by the camera, wherein the processed image has a histogram based on the target histograms, and using the processed image to determine state information associated with the objects. The processed image may be generated by processing the image captured by the camera using a histogram matching algorithm to generate the histogram of the processed image based on the target histograms.

Abstract: In one embodiment, a computing system may receive an uncompressed image from a camera. The computing system may generate a compressed image by performing a compression process on the uncompressed image, wherein a decompressed image may be generated as a byproduct during the compression process. The computing system may send the decompressed image to a machine-learning model that was trained using decompressed images. The computing system may generate, by the machine-learning model, an output based on the decompressed image. The computing system may provide operational instructions to a vehicle based on the output.

Abstract: In one embodiment, a method includes accessing a set of data points captured using a radar system of the vehicle. Each data point is associated with at least three measurements include a Doppler measurement, a range measurement, and an azimuth measurement in reference to the radar system. The method also includes clustering the set of data points into one or more first clusters based on a first pair of the three measurements associated with each of the data points; and clustering the set of data points into one or more second clusters based on a second pair of the three measurements associated with each of the data points. The second pair being different from the first pair of the three measurements.

Abstract: In one embodiment, a method includes receiving a first signal associated with a first multipath effect from a first radar installed on a vehicle at a first height, receiving a second signal associated with a second multipath effect from a second radar installed on the vehicle at a second height, wherein the first height and the second height are different, wherein a difference between the first height and the second height is configured to generate a mitigation of the first multipath effect and the second multipath effect, and wherein the first radar and the second radar have an overlapping field of view, and determining that a target exists in the overlapping field of view based on the first signal and the second signal.

Abstract: Examples disclosed herein may involve (i) obtaining a first layer of map data associated with sensor data capturing a geographical area, the first layer of map data comprising an aggregated overhead-view image of the geographical area, where the aggregated overhead-view image is generated from aggregated pixel values from a plurality of images associated with the geographical area, (ii) obtaining a second layer of map data, the second layer of map data comprising label data for the geographical area derived from the aggregated overhead-view image of the geographical area, and (iii) causing the first layer of map data and the second layer of map data to be presented to a user for curation of the label data.

Abstract: In one embodiment, a computing system of a vehicle may receive perception data associated with a scenario encountered by a vehicle while operating in an autonomous driving mode. The system may identify the scenario based at least on the perception data. The system may generate a performance metric associated with a vehicle navigation plan to navigate the vehicle in accordance with the identified scenario. In response to a determination that the performance metric associated with the vehicle navigation plan fails to satisfy one or more criteria for navigating the vehicle in accordance with the identified scenario, the system may trigger a disengagement operation related to disengaging the vehicle from the autonomous driving mode. The system may generate a disengagement record associated with the triggered disengagement operation. The disengagement record may include information associated with the identified scenario encountered by the vehicle related to the disengagement operation.

Abstract: In one embodiment, a computing system of a vehicle may receive vehicle driving data associated with a vehicle driving in an environment and detected environment data associated with the environment. The system may generate a reference trajectory of the vehicle driving in the environment based on the vehicle driving data. The system may determine driving constraints associated with the environment based on the detected environmental data. The system may generate a trajectory of the vehicle based on the driving constraints. The system may determine a difference in at least one parameter associated with the trajectory relative to at least one corresponding parameter associated with the reference trajectory. The system may adjust weight values associated with cost functions of the trajectory based on the difference between the at least one parameter associated with the trajectory and the corresponding parameter associated with the reference trajectory.

Abstract: Systems, methods, and non-transitory computer-readable media provide an adaptive gating mechanism for radar tracking initialization. Specifically, the radar system obtains measurement data of target points, and then determines, based on the measured position and dopplers of points in the first few scans, whether the doppler and displacement parameters satisfy an initialization constraint. When the initialization constraint is not satisfied, the radar system flags the respective cluster with an initialization flag, and adaptively uses the measured position and doppler of scanned points to determine the gating size for the next scan, instead of using a fixed gate size. When the initialization flag of the same cluster across a few consecutive scans satisfies a combination logic, the radar system determines that the tracking enters into the association stage, e.g., the radar system formally generates a track for the target points along a series of scans.

Abstract: Systems and methods include accessing streams of sensor data; constructing a corpus of seed sample data; initializing a first instance of a trained model using the corpus of seed sample data that: generates predictions of predicted sensor values; computing error values based on calculated differences between the actual sensor values and the predicted sensor values; transmitting the computed error values; initializing a second instance of the trained model based on an input of the corpus of the seed sample data, wherein the second instance of the trained model is identical to the first instance of the trained model, and wherein the second instance: generates inferences of predicted sensor values for each of the sensors based on the input of the corpus of seed sample data; reconstructing estimates of the actual sensor values based on a reconstruction computation with the parallel predicted sensor values and the error values.

Abstract: In one embodiment, a method includes configuring a radar transceiver to transmit a first number of radar pulses at a first pulse repetition frequency (PRF); and determining a first value corresponding to a first object based on a first radar data received in response to the first number of radar pulses. The first object is identified based on the first value being higher than a predetermined threshold value. The method also includes configuring the radar transceiver to transmit a second number of radar pulses at a second PRF that is higher than the first PRF; determining a second value of the first object based on a second radar data received in response to the second number of radar pulses; and associating the second value with information of the first object.

Abstract: In one embodiment, a method includes determining one or more transmission criteria for a radar system. The radar system having a number of antenna elements. The method also includes determining a configuration of a selected subset of the number of antenna elements satisfying the one or more transmission criteria; configuring a switching network to couple each antenna element of the selected subset of antenna elements to a corresponding receiving channel and a corresponding transmitting channel; and causing the radar system to transmit and receive signals using the selected subset of antenna elements.

Abstract: In one embodiment, a method includes accessing sensor data associated with an environment of the vehicle; identifying agents in the environment based on the sensor data; determining a probability distribution indicative of whether a preliminary trajectory of each of the agents intersects a potential travel space of the vehicle; generating a prioritization of the agents based on the probability distribution; allocating an amount of computing resources of the vehicle for analyzing each of one or more of the agents based on the prioritization and characteristics of the respective one or more agents; and predicting a trajectory of one or more of the agents according to the allocated computing resources. An accuracy of the prediction is proportional to the amount of allocated computing resources. The method also includes determining, based on the analysis of the agents, one or more operations for the vehicle to perform.

Abstract: Systems, methods, and non-transitory computer-readable media provide a blind online calibration mechanism to calibrate the position of the radar unit while the self-driving vehicle is in motion without the use of any map data. Specifically, a calibration system associated with the self-driving vehicle is configured to adjust the boresight angle of the radar within a calibration range and monitor the convergence or divergence pattern of the resulting clutter locations. The boresight angle of the radar unit may be progressively adjusted in static or dynamic degree increments until the convergence of the clutter curves is observed. In this way, without using any map data or factory settings, radar calibration can be conducted as often as needed while the vehicle is moving. Radar sensing and measurement accuracy is thus improved.

Abstract: A method includes receiving, by a computing system of a vehicle, a planned trajectory for the vehicle based on first sensor data associated with an external environment. The planned trajectory is generated by a primary computing system associated with a first sensor. The method further includes determining an environmental condition associated with the external environment. The environmental condition is included in second sensor data generated by a second sensor. The method further includes analyzing one or more parameters associated with the planned trajectory to determine whether the planned trajectory is based on the environmental condition, and, in response to determining that the primary computing system fails to base the planned trajectory upon the environmental condition, altering the one or more parameters associated with the planned trajectory to generate an altered planned trajectory. The method thus includes instructing the vehicle to execute the altered planned trajectory.

Abstract: Systems, methods, and non-transitory computer-readable media provide a cooling component including an underbody wheel well fan that is installed in proximity to the wheel well at each side of the front wheels. Specifically, an apparatus for vehicle radiator cooling control is provided, including a first suction component disposed in proximity to a first wheel well at a first side of a vehicle, a first tube component having a first end connected to the first suction component and a second end extended to a direction towards a back of the vehicle, and a first fan component connected to the second end of the first tube component.

Abstract: In one embodiment, a method includes receiving parameter data associated with trajectories captured using at least one sensor associated with one or more vehicles, determining a parameter distribution associated with movement parameters for the parameter data based on the trajectories, adjusting the parameter distribution for the parameter data associated with the movement parameters up to a respective threshold constraint, wherein the respective threshold constraint is based on at least one of the parameter data, generating driving simulations that are based on the adjusted parameter distribution for the parameter data, and evaluating an autonomous-driving system using the driving simulations.

Abstract: A method includes determining vehicle driving parameters associated with a driving behavior of a vehicle while in an autonomous driving mode. A plurality of electronic control units (ECUs) is interconnected to a vehicle bus for communicating instructions from electromechanical contacts to actuators to control the driving behavior. The computing unit may receive a control signal from one of the ECUs. The computing unit lacks knowledge of whether the control signal a valid. The method further includes analyzing vehicle driving parameters associated with the driving behavior to determine whether the vehicle driving parameters satisfy a driving behavior criteria. Subsequent to determining that the vehicle driving parameters fails to satisfy the driving behavior criteria, the method includes generating a signal that causes the electromechanical contacts to disconnect from the vehicle bus and disable communication of instructions between the ECUs and the vehicle bus.