A multi-objective optimization model of robot path planning under different scenarios

With the rapid development of logistics industry, path planning of mobile robots has been widely concerned. The main purpose of this paper is to solve the problem of path planning for autonomous mobile robots in static environment. The problem can be solved by determining a collision-free path that satisfies the selected criteria of shortest distance and path smoothness. The proposed multi-objective path planning model aims at scientific and reasonable path planning for mobile robots to increase the path smoothness and reduce the moving distance. Firstly, according to the characteristics of obstacles, two groups of different robot workspace scenarios were set from the degree and density of obstacles. Secondly, a multi-objective nonlinear programming model is established to maximize path smoothness and minimize path distance. Finally, different scenarios are simulated, and a fast non-dominated sorting genetic algorithm based on elite strategy (NSGAII) is used to solve the model. Simulation results show that the multi-objective path planning model established in this paper can generate an optimal feasible path even in a complex environment, so as to overcome the shortcomings of traditional methods such as grid method. Moreover, compared with recent path planning techniques, the proposed multi-objective evolutionary algorithm is highly competitive in path optimization.