Argo AI, LLC

To identify sources of data resulting from an execution flow in a robotic device such as an autonomous vehicle, an operating system receives sensor data from various sensors of the robotic device. For each sensor, the system generates a data log comprising an identifier of a first checkpoint associated with that sensor, as well as a first timestamp. The system performs an execution flow on the sensor data from that sensor. The system updates the data log to include an identifier and timestamp for one or more additional checkpoints during the execution flow. The system then fuses results, uses the fused data as an input for a decision process, and causes a component of the robotic device to take an action in response to an output of the decision process. The system may record the action, an action timestamp and the data logs for each sensor in a memory.

Systems and methods for operating an autonomous vehicle (AV) are provided. The method includes detecting one or more objects in an environment, predicting a first set of predicted object trajectories comprising one or more trajectories for each of the detected one or more objects, generating a plurality of candidate AV trajectories for the AV, scoring each of the candidate AV trajectories according to a cost function, using the scoring to select a final AV trajectory for execution, determining which of the predicted object trajectories affected the final AV trajectory and which did not do so, adding the predicted object trajectories that affected the final AV trajectory to a persisted prediction cache, excluding from the persisted prediction cache any predicted object trajectories that did not affect the final AV trajectory, and executing the final AV trajectory to cause the AV to move along the final AV trajectory.

Methods of identifying corner case simulation scenarios that are used to train an autonomous vehicle motion planning model are disclosed. A system selects a scene that includes data captured by one or more vehicles over a time period. The data includes one or more actors that the vehicle's sensors perceived over the time period in a real-world environment. The system selects a scene that includes a safety threshold violation, and it identifies the trajectory of an actor that participated in the violation. The system generates simulated scenes that alter the trajectory of the actor in the selected scene, selects simulated scenes that are more likely to occur in the real world and that may include safety threshold violations that go beyond any that may be found in the original scene, and uses the selected simulated scenes to train an autonomous vehicle motion planning model.

Disclosed herein are system, method, and computer program product aspects for enabling providing map updates to an autonomous vehicle (AV). The system can has four features that can (1) provide live map updates, (2) provide a mechanism to optimize map updates, (3) utilize AV sensors to provide local map updates and integrate into the broader map update system, and (4) provide a mechanism to provide low latency updates.

Systems, methods, and computer-readable media are disclosed for context aware verification for sensor pipelines. Autonomous vehicles (AVs) may include an extensive number of sensors to provide sufficient situational awareness to perception and control systems of the AV. For those systems to operate reliably, the data coming from the different sensors should be checked for integrity. To this end, the systems and methods described herein may use contextual clues to ensure that the data coming from the different the sensors is reliable.

Disclosed herein are system, method, and computer program product embodiments for an asymmetrical Autonomous Vehicle Systems (AVS). A backup AVS is implemented on a vehicle to serve as a failover system for one or more of the primary AVS components or processes (e.g., steering, braking, etc.). In this way, during primary AVS failures, the backup AVS can dynamically handle a subset of vehicle operations in various component configuration levels based on a desired mission level.

Disclosed herein are system, method, and computer program product embodiments for automated autonomous vehicle pose validation. An embodiment operates by generating a range image from a point cloud solution comprising a pose estimate for an autonomous vehicle. The embodiment queries the range image for predicted ranges and predicted class labels corresponding to lidar beams projected into the range image. The embodiment generates a vector of features from the range image. The embodiment compares a plurality of values to the vector of features using a binary classifier. The embodiment validates the autonomous vehicle pose based on the comparison of the plurality of values to the vector of features using the binary classifier

This document discloses system, method, and computer program product embodiments for operating an autonomous vehicle (AV). For example, the method includes performing the following operations by a muxing tool when AV is deployed within a particular geographic area in a real-world environment: receiving perception data that is representative of at least one actual object which is perceived while AV is deployed within the particular geographic area in a real-world environment; receiving simulation data that represents a simulated object that could be perceived by AV in the real-world environment and that was generated using a simulation scenario which is selected from a plurality of simulation scenarios based on at least one of the particular geographic area in which AV is currently located and a current operational state of AV; and generating augmented perception data by combining the simulation data with the perception data.

Disclosed herein are system, method, and computer program product embodiments for automated autonomous vehicle pose validation. An embodiment operates by generating a range image from a point cloud solution comprising a pose estimate for an autonomous vehicle. The embodiment queries the range image for predicted ranges and predicted class labels corresponding to lidar beams projected into the range image. The embodiment generates a vector of features from the range image. The embodiment compares a plurality of values to the vector of features using a binary classifier. The embodiment validates the autonomous vehicle pose based on the comparison of the plurality of values to the vector of features using the binary classifier.

Disclosed herein are system, method, and computer program product embodiments for clustering lane segments of a roadway in order to improve and simplify autonomous vehicle behavior testing. The approaches disclosed herein provide a hybrid methodology of dividing lane segments into hard features and soft features, and using a metric learning model trained in a supervised process on the entirety of lane segment features to cluster the lane segments based on the soft features. These clustered lane segments can then be assigned to what is termed as protolanes, where a single set of tests applied to a given protolane is considered valid across all of the lane segments assigned to the protolane.