Path Planning And Arm Control For Mobile Robot

In recent years,with the rapid popularity of the logistics,the intelligent storage and other industries,the mobile robots develop rapidly and the research of such kind of robots is also more and more popular.Based on the research background of the third generation service storage robot,in the process of moving the mobile robot to the designated location and then grasping the goods,the main research includes two aspects,one is the navigation path planning problem of mobile,on the other hand is the control problem in object grabbing.In this paper,the NAO robot produced by Aldebaran is used as the research object and the above two problems are studied and analyzed.The main content is as follows:First of all,for the navigation path planning research of NAO robot,the navigation map in unknown environment is built and the optimal path is applied in the path planning.As to the navigation map,the sonar sensor is adopted to detect the distances between robot,obstacles and terminal landmark.In order to get the number of obstacles,K-means algorithm is used to classify the data that obtained by sonar sensor.And a camera is used to distinguish the obstacles and the terminal landmark.Then,Q-learning algorithm is adopted to build navigation map for the robot.After that,simulate annealing(SA)algorithm and bi-directional parallel search strategy are applied to improve the performance of traditional artificial bee colony(ABC)algorithm.Then the improved algorithm is introduced to find optimal path of NAO robot in the off-line navigation map.The results show that the robot can avoid obstacles and effectively build the navigation map from the start point to the end landmark.It also finds an optimal path by the improved artificial bee colony(IABC)algorithm.Compared with ABC algorithm,the convergence speed of IABC algorithm is faster and it is not easy to fall into local optimum.Secondly,the dynamic modeling and control of NAO robot arm are researched.In this paper,the 2-DOF Euler-Lagrange dynamics model of NAO robot arm is established.Based on the dynamic model,two controllers are designed with or without considering the gravity of the arm,respectively.And the stability analysis for control system is included.Control system simulation results show that with gravity compensation of the PD controller can accelerate the response speed of system,and effectively improve the control accuracy of servo reference input.