Research On Motion Control And Motion Plan Of4WD Omni-directional Mobile Robot

As the developing of mobile robotic technologies, robots have vast applications in real life such outdoors exploration, reconnaissance on ocean floor, other planetary exploration, dangerous sites rescue, family service, and so on. And it plays more and more important role in our social production and life. Different from traditional wheeled robot, omnidirectional mobile robot can rotat with zero radiuses and move in any direction without changing any position and posture. With the advantages of its high motion flexibility and good maneuverability, the omnidirectional mobile robot has been widely noted and developed in recent years. Because the research on motion control may excavate the omnidirectional robot’s motion potentiality and develop its motion advantage as far as possible, as a result, the research is of great importantance both in theory and practice.After analyzing and summarizing the domestic and overseas studies, based on the adaptive control, backstepping, neural network, sliding mode control, brain emotional learning model, decoupling principle of multivariable system, harmony search algorithm and Lyapunov stability theory, the motion control of a four-wheel drived omnidirectional mobile robot (4WDOMR) is studied thoroughly. The main research work is as follows:(1) Trajectory tracking control of4WDOMR based on NDAP (Neural Dynamics with Adaptive Parameters)In order to resovle the speed jump existing in conventional backstepping tracking control for4WDOMR, we propose an adaptive controller based on the bioinspired neural dynamics model. Because of the smoothness and boundedness of the output from the bioinspired model, it produces a gradually varying tracking speed instead of the jumping speed, and the parameters are designed to avoid the control values exceeding its limits. And then, a parameter adaptive controller is presented to improve control performance. Simulation results of different paths and comparison with the conventional backstepping technique show that the approch is effective, and the system has a good performance with smooth outputs.(2) Velocity tracking control of4WDOMR based on IBEL (Improved Brain Emotional Learning)For there are mechanical difference and couple relationship in four wheels, even each motor has the optimal parameter, the robot may not be precisely controlled. A velocity controller is applied to motion control of4WDOMR. The controller is based on the improved brain-emotional-learning algorithm, whose learning steps are adapted online by fuzzy control approach. The improved brain emotional learning velocity controller (IBELC) can supply the additional control volumes for each wheel to improve the motion control accuracy of whole robot. The comparison between IBELC and normalized brain-emotional-learning-based intelligent controller is shown via simulation, and the results show that the controller has a good performance.(3) Trajectory tracking control of4WDOMR based on ASMCFR(Adaptive Sliding Mode Controll base on Filter and RBF neural network)For dynamic model of a4WDOMR usually contains parameter uncertainties, in addition, with the influence of exogenous disturbances, the traditional method for motion control has not good performance, the ASMCFR is presented. According to the variable structure control theory and Radial Basisi Function neural networks, combining the filter, the ASMCFR is applied to deal with the inherent buffeting with normal variable structure control method. To avoid the deteriorative control quality caused by change of parameters, a parameter adaptive law, which is used to estimate the variational parameters, is proposed. The contribution of ASMCFR in improving the control system performance is shown via simulation. The results show that this method has good tracking robustness and a high control precision, simple achievement and effectively eliminated buffeting.(4) Multi-motor decoupling control of4WDOMR based on RMSCD (Reference Model based on Separation of Control and Decoupling)The4WDOMR is a complicated nonlinear, strong coupling mechanical system, since the strong coupling phenomena exists between each wheel drive motor, it is difficult to obtain ideal control effect. In order to solve this problem, it is raised a decoupling method of consistency controller outputs for motor variable control based on dynamic analysis, through the four wheel robot dynamic analysis of driving torque of the four-wheel robot, the four wheel speed and state equation between drive torque are deduced to realize the four independent control of the motor. Compared with the traditional reference model decoupling control, the new controller meets both controlling and decoupling performance. Simulation results show that the method can reduce the interaction between the control variable with little errors(almost equal to10-5), and each motor can track their own input well, which shows good tracking decoupling effect.(5) Motion plan with energy consumption optimization based on IHS (Improved Harmony Search) for4WDOMRFor the traditional method, which is used to deal with the motion plan with energy consumption optimization for4WDOMR, usually has not a exact solution, a practical cost function as the total energy drawn from the batteries has been established, and then a motion plan model with energy consumption optimization solved with IHS algorithm is obtained. Various simulations are performed and the consumed energy is compared to the minimum-time plan and Kim’s approach. Simulation results reveal that the energy saving with the new approach is much more than that with minimum-time plan and Kim’s approach, the operational time of the4WDOMR with given batteries is lengthened. And compares with GA algorithm, PSO algorithm, HS algorithm, the IHS algorithm can not only improve the computational accuracy more effectively, but also has some advantages such as high algorithm convergence rate, fast convergence speed, strong global optimization ability, etc.At the end of this dissertation, the main research is summarized. It makes out the main innovations and research achievements, and also points out the problems and issues which need to further research.