Improved artificial bee colony algorithm based optimal navigation path for mobile robot

Considering path optimization problem on mobile robot, an experimental platform based on NAO robot is set up to build navigation map and get optimal path under unknown environment. Based on simultaneous localization and mapping (SLAM) theory, the sonar sensor of the NAO robot is adopted to detect distances between robot and obstacles and between robot and end landmark. K-means algorithm is used to classify the obstacle and calculate the accurate distances, and a sonar sensor and a camera are used to distinguish the obstacle and the landmark. Then, Q-learning algorithm is adopted to build navigation map for the robot. After that, bi-directional parallel search strategy and simulate annealing algorithm are proposed to improve traditional artificial bee colony algorithm, and the improved artificial bee colony algorithm is introduced to find optimal path in the navigation map. Finally, the experimental results show that the built navigation map based the robot can find an optimal path from the origin to the landmark avoiding the obstacles and a faster convergence speed can be obtained by the improved artificial bee colony algorithm.