Learning Enabled Fast Planning and Control in Dynamic Environments with Intermittent Information

• DOI: 10.48550/arxiv.2209.04534
• Publication Date: 2022-01-01
• Authors: Matthew Cleaveland; Esen Yel; Yiannis Kantaros; Insup Lee; Nicola Bezzo
• Keywords: FOS: Computer and information sciences; Computer Science - Robotics; 0209 industrial biotechnology; FOS: Electrical engineering, electronic engineering, information engineering; Systems and Control (eess.SY); 02 engineering and technology; Electrical Engineering and Systems Science - Systems and Control; Robotics (cs.RO)

This paper addresses a safe planning and control problem for mobile robots operating in communication- sensor-limited dynamic environments. In this case the cannot sense objects around them must instead rely on intermittent, external information about environment, as e.g., underwater applications. The challenge is that plan using only stale data, while accounting any noise data or uncertainty environment. To address we propose compositional technique which leverages neural networks to quickly robot through crowded environments intermittent information. Specifically, our tool uses reachability analysis potential fields train network capable of generating actions. We demonstrate both simulation with an vehicle crossing shipping channel real experiments ground vehicles