A novel intelligent hybrid controller for optimal path navigation in unknown environments

This paper aims to solve the path navigation complexity for quad wheel robots in unknown environments by introducing one Hybrid Particle Swarm Optimization Bug (HPSO-BUG) controller. In the path navigation process, HPSO-BUG follows an M-Line between the source and target location like Bug2 controller. In the obstacle avoidance process, HPSO-BUG determines the obstacle encounter point and then, it searches for the best obstacle leaving point. HPSO-BUG seeks target from the obstacle leaving point with the help of a suitable fitness function which was taken from fitness values of randomly generated data particles with their velocities. Particle’s best value has been updated after several iterations and then fitted into the wheeled robot path navigation problem. In the simulation, the proposed controller exhibited an average 4.18% deviation in terms of travel distance and an average 4.94% deviation in terms of travel time. In real-time, the proposed controller was found to have an average 88.75% success rate and has respectively 0.25%, 1.25%, 0.89%, and 14.98% less deviation in terms of path length and 15.07%, 25.34%, 21.21%, and 49.89% less deviation in terms of travel time compared to PSO-tuned FNN, BA, FPA, and IWO.