Human-aware motion planning based on search and sampling approach

Abstract

With the improvement on various technical aspects of robotics, autonomous robotics gradually immerges into our daily life and the chance of human-robot interaction rises especially in dynamic environments. To carry out different tasks and in many cases collaborating with human, effective motion planning must be devised. Besides of yielding real-time and cost effective trajectories, the social aware motion planning approach must also respect social norms in terms of avoidance of personal space. We have therefore propose an alternative and novel human-aware navigation integrating dynamic social costmap together with an anytime dynamic RRT* planner to achieve viable social aware motion planning. The experiments on several environments verify the practicality of such autonomous social aware navigation strategy.