Sliding Autonomy for UAV Path-Planning: Adding New Dimensions to Autonomy Management

Abstract

Increased use of autonomy also increases the need for humans to interact with or manage autonomy. We propose a new variation of sliding autonomy useful for planning problems over a spatial region. With this approach, the user can influence the behavior of the autonomous system via spatial constraints and temporal constraints. We present a set of user interface designs to implement sliding autonomy for Unmanned Aerial Vehicle path-planning to support Wilderness Search and Rescue. Interactivities along these new dimensions allow the user to allocate degrees of authority and flexibility to the robot's algorithms. We evaluate the usefulness of the approach against manual and simple pattern path-planning methods with a user study. Results show that the sliding autonomy approach performs significantly better than the other two methods without increasing the users' mental workload, and the performance of the human-autonomy team outperforms either human or autonomy working alone.