Trajectory Planning for Robots in Dynamic Human Environments

This paper present a trajectory planning algorithmfor a robot operating in dynamic human environments.Environments such as pedestrian streets, hospital corridors andtrain stations. We formulate the problem as planning a minimalcost trajectory through a potential field, defined from theperceived position and motion of persons in the environment. ARapidly-exploring Random Tree (RRT) algorithm is proposedas a solution to the planning problem.A new method for selecting the best trajectory in the RRT,according to the cost of traversing a potential field, is presented.The RRT expansion is enhanced to direct the search andaccount for the kinodynamic robot constraints. Compared tostandard RRT, the algorithm proposed here find the robotcontrol input that will drive the robot towards a new sampledpoint in the configuration space. The effect of the input issimulated, to add a reachable vertex to the tree.Instead of executing a whole trajectory, when planned, thealgorithm uses an Model Predictive Control (MPC) approach,where only a short segment of the trajectory is executed whilea new iteration of the RRT is done.The planning algorithm is demonstrated in a simulatedpedestrian street environment