Time-Optimized Contextual Information Flow on Unmanned Vehicles

Nowadays, the domain of robotics experiences a significant growth. We focus on Unmanned Vehicles intended for the air, sea and ground (UxV). Such devices are typically equipped with numerous sensors that detect contextual parameters from the broader environment, e.g., obstacles, temperature. Sensors report their findings (telemetry) to other systems, e.g., back-end systems, that further process the captured information while the UxV receives control inputs, such as navigation commands from other systems, e.g., commanding stations. We investigate a framework that monitors network condition parameters including signal strength and prioritizes the transmission of control messages and telemetry. This framework relies on the Theory of Optimal Stopping to assess in real-time the trade-off between the delivery of the messages and the network quality statistics and optimally schedules critical information delivery to back-end systems.