Abstract: Data management of vehicle data for communication to a remote location is provided, including obtaining one or more sensor data feeds from one or more onboard sensors of a vehicle, generating a forecast network condition for a network with which the vehicle is in operative communication, the forecast network condition representing predicted network resources the network is predicted to provide to the vehicle, providing the forecast network condition to a compression engine, and compressing the one or more sensor data feeds into a compressed data feed for communication over the network to a remote location, wherein the compressed data feed is compressed based on the forecast network condition.

Abstract: Data management of vehicle data for communication to a remote location is provided, including obtaining one or more sensor data feeds from one or more onboard sensors of a vehicle, generating a forecast network condition for a network with which the vehicle is in operative communication, the forecast network condition representing predicted network resources the network is predicted to provide to the vehicle, providing the forecast network condition to a compression engine, and compressing the one or more sensor data feeds into a compressed data feed for communication over the network to a remote location, wherein the compressed data feed is compressed based on the forecast network condition.

Abstract: Predictive switching between autonomous control of a vehicle by an autonomous driving system and a remote pilot connected to the vehicle by a network. The predictive switching may include determining a route of the vehicle and identifying one or more suspect areas along the route in which the autonomous diving system is suspected of difficulty in autonomous control of the vehicle. Ahead of encountering the suspect areas, a remote driver may be assigned to the vehicle to establish communication with the vehicle and receive sensor data. This may allow for the remote pilot to monitor the vehicle including any projected autonomous operations. The remote pilot may intervene to assume control of the vehicle in the suspect area.

Abstract: Processing sensor data from a plurality of sensors monitoring an exterior environment of a remotely controlled vehicle. The sensors generate a corresponding respective plurality of sensor data feeds. A compositor for combining the plurality of sensor data feeds into a composite data stream is provided. A remote pilot terminal is operative to receive the composite data stream and render a representation of the exterior environment of the remotely controlled vehicle to a remote pilot. A pilot monitor determines an area of focus of the remote pilot, and the area of focus is provided to the compositor for use in differentiating the plurality of data feeds in the composite data stream. The representation of the exterior environment at the remote operator terminal is based on the area of focus of the remote pilot.

Abstract: Systems and methods for redundant braking and/or of remotely piloted vehicles are described. A redundant braking system includes a first onboard computing unit configured to receive sync data regularly from a main computing unit, receive heartbeat signal through a communication network from a remote pilot control unit, determine network failure if the heartbeat signal is not received for more than a threshold period of time, and generate a braking command. The system includes arrangements to activate brakes of the vehicle based on the braking command received from the local computing unit. The system may also generate a steering command to change the lane of the vehicle and park at a safe spot.

Abstract: A remote pilot and driver assist system and methods is disclosed. In some embodiments, the remote pilot and driver assist system and methods may be used to safely navigate a vehicle to a desired destination. The remote pilot and driver assist system and methods may include a driver assist application and a data store running on an application server and wherein the application server is in communication with one or more vehicles, one or more local users or drivers, and one or more remote pilots using their remote pilot terminals.

Abstract: Systems and methods are directed to using machine learning in determining vehicle status and/or rider status associated with a service trip. In one example, a computer-implemented method for providing a vehicle trip check includes obtaining sensor data from one or more sensors positioned within a cabin of a vehicle, the sensor data being descriptive of objects located within the cabin of the vehicle. The method further includes inputting the sensor data to a machine-learned trip check model, and includes receiving, as an output of the machine-learned trip check model, trip check analysis data. The method further includes determining, based on the trip check analysis data, that the trip check analysis data meets one or more predetermined criteria. The method further includes in response to determining that the trip check analysis data meets the one or more predetermined criteria, generating a trip control signal associated with operation of the vehicle.

Abstract: Systems and methods are directed to using machine learning to determine passenger ride experience. In one example, a computer-implemented method includes receiving, by a computing system comprising one or more computing devices, sensor data from one or more sensors positioned within a cabin of a vehicle, the sensor data being descriptive of one or more passengers located within the cabin of the vehicle. The method further includes inputting, by the computing system, the sensor data to a machine-learned ride experience model and receiving, as an output of the machine-learned ride experience model, ride experience data including ride experience events detected from the sensor data and a classification for each detected ride experience event according to a ride experience rating. The method further includes determining, by the computing system and based on the ride experience rating for each detected ride experience event, a ride experience control signal associated with operation of the vehicle.

Abstract: Systems and methods are directed to a task management platform for managing autonomous vehicles. In one example, a computer-implemented method for managing a fleet of vehicles includes obtaining, by a computing system comprising one or more computing devices, vehicle state data indicative of vehicle state from one or more vehicles among a fleet of vehicles. The method further includes determining, by the computing system, if a vehicle event has occurred based at least in part on the vehicle state data. The method further includes in response to determining a vehicle event has occurred, generating, by the computing system, one or more tasks for resolution of the vehicle event, wherein the one or more tasks include data associated with the one or more vehicles and the vehicle event. The method further includes providing, by the computing system, data associated with the one or more tasks to a remote operator computing system.