Research On Path Planning And Handing Based On Multi-NAO Robots

In recent years, the field of mobile robotics, the promotion of function and applications of the single robot have made tremendous progress, but at the same time multi-robot’s robustness, low cost and high efficiency are more excellent, so that the multi-robot coordination control gradually become a hot topic.Multi-robot path planning and handing are representative subject of multi-robot systems in the field, at this stage, the theoretical study of path planning and handing are more in simulation stage, but lack of actual research. This paper discusses multi-robot path planning and collaboration handling the two contents were studied based on the French company Aldebaran Robotics NAO humanoid intelligent robot experimental platform.This paper introduces multi-robot systems and their current status to explain the path planning and handling existing research, and analysis the current situation at home and abroad in both subject areas and the application of part of the research results.Orientation is a critical issue for robot path planning and collaboration handling. This paper study target-recognition and localization base on NAO robot vision system. According to the characteristics of the image to be captured and target identification, uses suitable image processing methods and positioning model, get the position of target identification in the NAO robot coordinate system, and using the positioning experiments to verify the feasibility of targeting methods.Artificial potential field method has the advantages of simple function, high operation efficiency, but easy to fall into local minima. This paper improves artificial potential field and verified it by using Matlab simulation. Then drawing lessons from simulation of the use of the angle parameters, design the rules for obstacle avoidance of NAO robot, so NAO robot can active avoidance while one or two obstacles exist between target with its own.Collaboration handing, after obstacle avoidance of NAO robot, including the position adjusting and target handling. In the position adjustment process, according to the results of active avoidance, using NAO robot vision system for handling multiple recognition target to achieve position adjustment; the target the handling process, by introducing NAO robot gait controller, through the precise control gait NAO robot to achieve formation keeping.The last part of the experiment, by repeated experiments, active avoidance, position adjustment and target handling, solve the problem appeared in the experiment, and get the best gait parameter table. Finally, the algorithm and program of Multi-NAO robot path planning and handing are verified by experiments.