An intelligent optimize path planner for efficient mobile robot path planning in a complex terrain

Abstract

Path planning is an essential necessity for the proper functioning of mobile robot in a complex terrain. Conventional approaches face different challenges such as balancing exploration and exploitation ability, premature convergence, weak searching ability, and longer path length. To overcome these flaws, an Intelligent Modified Particle Swarm Optimization approach with a different strategy is proposed. Firstly, a velocity regularized strategy based on regularized coefficients has been applied to balance the exploration and exploitation ability. Secondly, a neighborhood search strategy based on reward value and utilization probability has been employed, enriching search behaviors and avoiding premature convergence. Finally, a path smoothness principle based on hypocycloid curves has been used to smooth the sharp turns. The comparative analysis conducted in four different terrains with different complexity. Different performance indices are being measured to validate the effectiveness of the proposed approach. The outcome acquired from different terrains indicates that the proposed approach outperforms the GA-PSO, Advance PSO, FACO, and other conventional approaches with a maximum improvement (%) of 17.59% in path length and 76.66% in convergence rate.