Automatic programming of behavior-based robots using reinforcement learning
Artificial Intelligence
Strongly typed genetic programming
Evolutionary Computation
Evolution of homing navigation in a real mobile robot
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Hi-index | 0.00 |
In this paper, an intelligent control of a mobile robot in order to inspect damage in welding joint of various steel structures is presented. In the mobile robot, a drive method with caterpillar of equipped stronger magnet coil is employed to navigate on wall of a welding structure. And, a few of photo sensors and a CCD camera are equipped in the robot for recognizing joint line of welding and inspecting welding surface. Here, a genetic algorithm for predicting damage information in the joint line of welding is proposed, and it is possible that recognized map and photograph on the joint line of welding be reproduced on the screen for remote control. Depending on the experiments, reliability of the genetic algorithm is confirmed in the reality of the welding.