Abstract: Systems and methods for power and thermal management of autonomous vehicles are provided. In one example embodiment, a computing system includes processor(s) and one or more tangible, non-transitory, computer readable media that collectively store instructions that when executed by the processor(s) cause the computing system to perform operations. The operations include obtaining data associated with an autonomous vehicle. The operations include identifying one or more vehicle parameters associated with the autonomous vehicle based at least in part on the data associated with the autonomous vehicle. The operations include determining a modification to one or more operating characteristics of one or more systems onboard the autonomous vehicle based at least in part on the one or more vehicle parameters.

Abstract: An autonomous robot is provided. In one example embodiment, an autonomous robot can include a main body including one or more compartments. The one or more compartments can be configured to provide support for transporting an item. The autonomous robot can include a mobility assembly affixed to the main body and a sensor configured to obtain sensor data associated with a surrounding environment of the autonomous robot. The autonomous robot can include a computing system configured to plan a motion of the autonomous robot based at least in part on the sensor data. The computing system can be operably connected to the mobility assembly for controlling a motion of the autonomous robot. The autonomous robot can include a coupling assembly configured to temporarily secure the autonomous robot to an autonomous vehicle. The autonomous robot can include a power system and a ventilation system that can interface with the autonomous vehicle.

Abstract: Systems and methods for controlling an autonomous vehicle in response to vehicle instructions from a remote computing system are provided. In one example embodiment, a computer-implemented method includes controlling a first autonomous vehicle to provide a vehicle service, the first autonomous vehicle being associated with a first convoy that includes one or more second autonomous vehicles. The method includes receiving one or more communications from a remote computing system associated with a third-party entity, the one or more communications including one or more vehicle instructions. The method includes coordinating with the one or more second autonomous vehicles to determine one or more vehicle actions to perform in response to receiving the one or more vehicle instructions from the third-party entity. The method includes controlling the first autonomous vehicle to implement the one or more vehicle actions.

Abstract: Systems, methods, tangible non-transitory computer-readable media, and devices for detecting objects are provided. For example, the disclosed technology can obtain a representation of sensor data associated with an environment surrounding a vehicle. Further, the sensor data can include sensor data points. A point classification and point property estimation can be determined for each of the sensor data points and a portion of the sensor data points can be clustered into an object instance based on the point classification and point property estimation for each of the sensor data points. A collection of point classifications and point property estimations can be determined for the portion of the sensor data points clustered into the object instance. Furthermore, object instance property estimations for the object instance can be determined based on the collection of point classifications and point property estimations for the portion of the sensor data points clustered into the object instance.

Abstract: Systems and methods for providing an autonomous vehicle service are provided. A method can include obtaining data indicative of a service associated with a user, and obtaining data indicative of a transportation of an autonomous robot. The method can include determining one or more service configurations for the service. The method can include obtaining data indicative of a selected service configuration from among the one or more service configurations, and determining a service assignment for an autonomous vehicle based at least in part on the selected service configuration. The service assignment can indicate that the autonomous vehicle is to transport the user from the service-start location to the service-end location. The method can include communicating data indicative of the service assignment to the autonomous vehicle to perform the service.

Abstract: Systems and methods for predicting object motion and controlling autonomous vehicles are provided. In one example embodiment, a computer implemented method includes obtaining state data indicative of at least a current or a past state of an object that is within a surrounding environment of an autonomous vehicle. The method includes obtaining data associated with a geographic area in which the object is located. The method includes generating a combined data set associated with the object based at least in part on a fusion of the state data and the data associated with the geographic area in which the object is located. The method includes obtaining data indicative of a machine-learned model. The method includes inputting the combined data set into the machine-learned model. The method includes receiving an output from the machine-learned model. The output can be indicative of a plurality of predicted trajectories of the object.

Abstract: Systems, methods, tangible non-transitory computer-readable media, and devices for operating an autonomous vehicle are provided. For example, the disclosed technology can include receiving state data that includes information associated with states of an autonomous vehicle and an environment external to the autonomous vehicle. Responsive to the state data satisfying vehicle stoppage criteria, vehicle stoppage conditions can be determined to have occurred. A severity level of the vehicle stoppage conditions can be selected from a plurality of available severity levels respectively associated with a plurality of different sets of constraints. A motion plan can be generated based on the state data. The motion plan can include information associated with locations for the autonomous vehicle to traverse at time intervals corresponding to the locations. Further, the locations can include a current location of the autonomous vehicle and a destination location at which the autonomous vehicle stops traveling.

Abstract: The present disclosure provides systems and methods for training probabilistic object motion prediction models using non-differentiable representations of prior knowledge. As one example, object motion prediction models can be used by autonomous vehicles to probabilistically predict the future location(s) of observed objects (e.g., other vehicles, bicyclists, pedestrians, etc.). For example, such models can output a probability distribution that provides a distribution of probabilities for the future location(s) of each object at one or more future times. Aspects of the present disclosure enable these models to be trained using non-differentiable prior knowledge about motion of objects within the autonomous vehicle's environment such as, for example, prior knowledge about lane or road geometry or topology and/or traffic information such as current traffic control states (e.g., traffic light status).

Abstract: In some examples, a control system for a vehicle can detect a set of ultrasonic signals generated by a mobile computing device of a user. Additionally, the control system can determine a pin code from the set of ultrasonic signals. Moreover, the control system can perform one or more vehicle operations to initiate fulfillment of a transport request that is associated with the determined pin code, upon the user being determined to be within a given proximity distance of the vehicle.

Abstract: Systems and methods for autonomous vehicle motion planning are provided. Sensor data describing an environment of an autonomous vehicle and an initial travel path for the autonomous vehicle through the environment can be obtained. A number of trajectories for the autonomous vehicle are generated based on the sensor data and the initial travel path. The trajectories can be evaluated by generating a number of costs for each trajectory. The costs can include a safety cost and a total cost. Each cost is generated by a cost function created in accordance with a number of relational propositions defining desired relationships between the number of costs. A subset of trajectories can be determined from the trajectories based on the safety cost and an optimal trajectory can be determined from the subset of trajectories based on the total cost. The autonomous vehicle can control its motion in accordance with the optimal trajectory.

Abstract: The present disclosure provides autonomous vehicle systems and methods that include or otherwise leverage a machine-learned yield model. In particular, the machine-learned yield model can be trained or otherwise configured to receive and process feature data descriptive of objects perceived by the autonomous vehicle and/or the surrounding environment and, in response to receipt of the feature data, provide yield decisions for the autonomous vehicle relative to the objects. For example, a yield decision for a first object can describe a yield behavior for the autonomous vehicle relative to the first object (e.g., yield to the first object or do not yield to the first object). Example objects include traffic signals, additional vehicles, or other objects. The motion of the autonomous vehicle can be controlled in accordance with the yield decisions provided by the machine-learned yield model.

Abstract: Vehicle control systems can include one or more location sensors, an energy storage device, one or more charge sensors and one or more vehicle computing devices. The location sensor(s) can determine a current location of a vehicle, while the charge sensor(s) can determine a current state of charge of an energy storage device that can be located onboard the vehicle to provide operating power for one or more vehicle systems. The vehicle computing device(s) can communicate the current location of the vehicle and current state of charge of the energy storage device to a remote computing device, receive from the remote computing device a charging control signal determined, at least in part, from the current location of the vehicle and the current state of charge of the energy storage device, and control charging of the energy storage device in accordance with the charging control signal.

Abstract: A computer-implemented method for localizing a vehicle can include accessing, by a computing system comprising one or more computing devices, a machine-learned retrieval model that has been trained using a ground truth dataset comprising a plurality of pre-localized sensor observations. Each of the plurality of pre-localized sensor observations has a predetermined pose value associated with a previously obtained sensor reading representation. The method also includes obtaining, by the computing system, a current sensor reading representation obtained by one or more sensors located at the vehicle. The method also includes inputting, by the computing system, the current sensor reading representation into the machine-learned retrieval model.

Abstract: Systems and methods are directed to an autonomy computing system for an autonomous vehicle. The autonomy computing system can include functional circuits associated with a first compute function of the autonomous vehicle. Each of the functional circuits can be configured to obtain sensor data that describes one or more aspects of an environment external to the autonomous vehicle at a current time. Each of the functional circuits can be configured to generate, over a time period and based on the sensor data, a respective output according to the specified order. The autonomy computing system can include monitoring circuits configured to evaluate an output consistency of the respective outputs, and in response to detecting an output inconsistency between two or more of the respective outputs, generate data indicative of a detected anomaly associated with the first autonomous compute function.

Abstract: Systems and methods for controlling a failover response of an autonomous vehicle are provided. In one example embodiment, a method includes determining, by one or more computing devices on-board an autonomous vehicle, an operational mode of the autonomous vehicle. The autonomous vehicle is configured to operate in at least a first operational mode in which a human driver is present in the autonomous vehicle and a second operational mode in which the human driver is not present in the autonomous vehicle. The method includes detecting a triggering event associated with the autonomous vehicle. The method includes determining actions to be performed by the autonomous vehicle in response to the triggering event based at least in part on the operational mode. The method includes providing one or more control signals to one or more of the systems on-board the autonomous vehicle to perform the one or more actions in response to the triggering event.

Abstract: The present disclosure provides a sensor cleaning system that cleans one or more sensors of an autonomous vehicle. Each sensor can have one or more corresponding sensor cleaning units that are configured to clean such sensor using a fluid (e.g., a gas or a liquid). Thus, the sensor cleaning system can include both a gas cleaning system and a liquid cleaning system. According to one aspect, the sensor cleaning system can provide individualized cleaning of the autonomous vehicle sensors. According to another aspect, a liquid cleaning system can be pressurized or otherwise powered by the gas cleaning system or other gas system.

Abstract: Generally, the disclosed systems and methods utilize multi-task machine-learned models for object intention determination in autonomous driving applications. For example, a computing system can receive sensor data obtained relative to an autonomous vehicle and map data associated with a surrounding geographic environment of the autonomous vehicle. The sensor data and map data can be provided as input to a machine-learned intent model. The computing system can receive a jointly determined prediction from the machine-learned intent model for multiple outputs including at least one detection output indicative of one or more objects detected within the surrounding environment of the autonomous vehicle, a first corresponding forecasting output descriptive of a trajectory indicative of an expected path of the one or more objects towards a goal location, and/or a second corresponding forecasting output descriptive of a discrete behavior intention determined from a predefined group of possible behavior intentions.

Abstract: Various examples are directed to systems and methods for operating a vehicle comprising a tractor and a trailer attached for pulling behind the tractor. A center-of-mass system may determine a mass of the trailer and a tractor understeer. The center-of-mass system may determine the tractor understeer using steering input data describing a steering angle of the tractor and yaw data describing a yaw of the tractor. The center-of-mass system may determine a load center of mass using the tractor understeer and a mass of the trailer. The center-of-mass system may further determine that the load center of mass transgresses a center-of-mass threshold and send an alert message indicating that the load transgresses the load center-of-mass threshold.

Abstract: The present disclosure provides systems and methods that combine physics-based systems with machine learning to generate synthetic LiDAR data that accurately mimics a real-world LiDAR sensor system. In particular, aspects of the present disclosure combine physics-based rendering with machine-learned models such as deep neural networks to simulate both the geometry and intensity of the LiDAR sensor. As one example, a physics-based ray casting approach can be used on a three-dimensional map of an environment to generate an initial three-dimensional point cloud that mimics LiDAR data. According to an aspect of the present disclosure, a machine-learned model can predict one or more dropout probabilities for one or more of the points in the initial three-dimensional point cloud, thereby generating an adjusted three-dimensional point cloud which more realistically simulates real-world LiDAR data.

Abstract: Systems and methods for automatically adjusting the interior cabin of an autonomous vehicle are provided. In one example embodiment, an autonomous vehicle can include a main body including a floor and a ceiling that at least partially define an interior cabin of the autonomous vehicle. The autonomous vehicle can include a partition wall that is movable within the interior cabin of the autonomous vehicle. The partition wall can extend between the floor to the ceiling of the main body. The autonomous vehicle can include a computing system configured to receive data indicative of one or more service assignments associated with the autonomous vehicle and to adjust a position of the partition wall within the interior cabin based at least in part on the one or more service assignments.