Robot motion planning using Predictive trajectory of moving obstacles

The presence of robots among humans and their ability to perform certain tasks has exhibited remarkable growth in recent years. As a result, there is an essential need for powerful path planning algorithms to find feasible, safe, and collision-free paths for robots in the presence of humans, considering other factors and the environment. This paper presents an algorithm for the motion planning of an agent (robot) among pedestrians based on future decisions. In each step, the robot selects its next velocity and subsequently its next position by predicting the movements of neighboring pedestrians. Based on the predicted future positions of pedestrians , a force is applied to the agent to determine its next velocity and position. Experimental results demonstrate the improvement of the proposed method in terms of the time optimality and safety (reducing number of collisions among robot and pedestrians) of the produced path compared to the scenario where prediction is not utilized.