A shared control method for obstacle avoidance with mobile robots and its interaction with communication delay

Abstract

Teleoperation allows human operators to safely extend themselves to remote environments that are typically difficult or dangerous to access. The remote environments are often unstructured (i.e. not having clear roads or paths to follow) and only accessible by wireless communication (introducing factors such as degraded signals and communication delay). Teleoperated driving under these conditions can result in slow operation speeds and unintended collisions with obstacles. Automating portions of the teleoperation task can help mitigate some of the negative effects of wireless communication. Shared control is used to combine inputs from the human teleoperator and automation. This work presents a new model predictive control based shared control method. We introduce a new representation for obstacle free regions that works well with unstructured robot environments and allows for an model predictive control problem formulation that can be solved rapidly. The shared control method is implemented in a robot simulator and tested with human subjects. Two user studies involving a search task with a mobile robot evaluate the effectiveness of the shared control method and explore its interaction with factors such as communication delay and input interface style. Communication delay is found to have the largest magnitude effect on performance and safety measures. Results demonstrate that the shared control method can improve both performance and safety when delays are present.