Efficient and Robust Robot Path Planning Based on Novel Wolf Pack Algorithm

Abstract

In the context of the wave of intelligence, the effective identification of the environment and reasonable planning of the path are the major goals of visual robots. The quality of the path planning determines the efficiency of the completion of the rest of the tasks, and even the success of the task. In this study, the scheme is designed from 3 aspects: 1. the robot vision system to sense the information around is designed. This study applies the visual SLAM algorithm of the graph optimization framework with the hardware improvement to make the sensing performance better; 2. the novel gray wolf pack algorithm is designed. The algorithm has strong convergence performance, and has characteristics of the less parameters and easy implementation. Further, the parameters are optimized to make the performance better than the traditional ones; 3. the novel path planning system is implemented. In the experiment section, the testing on moving time and path length are both tested with the comparison of the traditional methods, it is shown that the proposed model has the better performance.