Abstract: In various examples, physical sensor data may be generated by a vehicle in a real-world environment. The physical sensor data may be used to train deep neural networks (DNNs). The DNNs may then be tested in a simulated environment—in some examples using hardware configured for installation in a vehicle to execute an autonomous driving software stack—to control a virtual vehicle in the simulated environment or to otherwise test, verify, or validate the outputs of the DNNs. Prior to use by the DNNs, virtual sensor data generated by virtual sensors within the simulated environment may be encoded to a format consistent with the format of the physical sensor data generated by the vehicle.

Abstract: In various examples, physical sensor data may be generated by a vehicle in a real-world environment. The physical sensor data may be used to train deep neural networks (DNNs). The DNNs may then be tested in a simulated environment—in some examples using hardware configured for installation in a vehicle to execute an autonomous driving software stack—to control a virtual vehicle in the simulated environment or to otherwise test, verify, or validate the outputs of the DNNs. Prior to use by the DNNs, virtual sensor data generated by virtual sensors within the simulated environment may be encoded to a format consistent with the format of the physical sensor data generated by the vehicle.

Abstract: In various examples, physical sensor data may be generated by a vehicle in a real-world environment. The physical sensor data may be used to train deep neural networks (DNNs). The DNNs may then be tested in a simulated environment—in some examples using hardware configured for installation in a vehicle to execute an autonomous driving software stack—to control a virtual vehicle in the simulated environment or to otherwise test, verify, or validate the outputs of the DNNs. Prior to use by the DNNs, virtual sensor data generated by virtual sensors within the simulated environment may be encoded to a format consistent with the format of the physical sensor data generated by the vehicle.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for routing video feeds from different types of cameras to a server over a data communications network. In one aspect, a method performed by a camera intercept device includes identifying at least first and second cameras; configuring, using a first configuration procedure, the first camera to transmit a first video feed to the camera intercept device; configuring, using a second configuration procedure different from the first configuration procedure, the second camera to transmit a second video feed to the camera intercept device; and routing the first and second video feeds to a server over a data communications network using the wireless communications module.

Abstract: A mobile robot system for video teleconferencing system is described herein. A robot can be operable by a remote user. Via the robot, a remote user can interact within a local user's environment to provide telepresence capabilities. For example, the robot can be deployed on a horizontal surface, such as a table or desktop. The robot can include a microcontroller, a drive unit, and interface to a consumer device, such as a mobile device. The drive unit can include two or more motors for providing motion capabilities. The microcontroller can be wired or communicatively coupled to the consumer device. In general, the consumer device may be a mobile phone or a tablet computer where processing power and wireless or other capabilities of these devices can be utilized by the system. The system can be based on a networking protocol providing multi-party data exchanges.