Study On The Control Of Trajectory Tracking And Flocking Motion For The Autonomous Navigation Of Mobile Robots

With the development of science and technology, autonomous mobile robotshave attracted many researchers from the control and artificial intelligence community.Recently, modern mobile robots have entered a new research stage, and have beenapplied to many fields, such as industry, agriculture, national defence, service,scientific research and so on. The intelligent mobile robots are capable of performingsome disgusting, heavy or harsh tasks historically assigned to human beings. In thisdissertation, we mainly concentrate on two main control techniques of trajectorytracking and flocking motion for the autonomous navigation of wheeled mobile robots.In the first part, we focus on the modeling and trajectory tracking control of wheeledmobile robot with non-ideal constraint. Moreover, the novel trajectory trackingcontrol algorithms based on kinematic and dynamic model respectively are present toimprove the accuracy of trajectory tracking for the wheeled mobile robot. In thesecond part, we investigate the problem of autonoumous navigation and flockingmotion control in unknown condition. In order to explore new approaches ofautonomous navigation and promote the intelligence of half an autonomous mobilerobot, we introduce the virtual sub-goal and tabu search algorithm to path planningtechniques of mobile robot, and propose the corresponding algorithms of autonomousnavigation and flocking motion control for mobile robots. The main research workand contributions of the dissertation are summarized as follows:1. For wheeled mobile robot with two independent driven wheel angularvelocities and non-ideal constraint, the kinematic model, dynamic model and drivenmodel of mobile robot are established for the research purposes and foundation oftrajectory tracking control, autonomous navigation and flocking motion control inunknown environment.2. A robot kinematic model with two independent driven wheel angular velocitiesas control inputs is obtained by analyzing the trajectory tracking problem for wheeledmobile robot whose mass centers are not coincidental with the geometrical centers.For the unknown wheel radius and the distance between the two driving wheels, anadaptive sliding-mode tracking control law is proposed by employing backsteppingmethod and sliding-mode control. The global asymptotically stability of the closedloop system is guaranteed by Lyapunov stability theory.3. In order to solve the trajectory tracking problem for the dynamic model of mobile robot with non-ideal constraint, an adaptive integral sliding-mode control lawis proposed in this chapter. A virtual control variable is firstly introduced according tothe kinematic model. Then an adaptive sliding-mode controller is designed byemploying backstepping technique and intergral sliding-mode control algorithm,which can realize the trajectory tracking requirements with strong robustness. Finally,the effectiveness and correctiveness of the proposed control law are demonstrated bysimulation examples.4. On the basis of the dynamic model for wheeled mobile robot, the driven motordynamics is taken into account further to improve the performance of trajectorytracking. By employing the voltage of motor as control input, an adaptive trajectorytracking control law for mobile robot system with uncertain inertia parameters,uncertain structure parameters and external disturbance is presented for this case toensure robustness to the uncertainty of model parameters. The proposed control lawcan guarantee that the mobile robot will track the desired trajectory accurately.5. Due to the lack of predictability by using reactive navigation algorithm inunknown condition, an active routing navigation algorithm based on sub-goal isdesigned for mobile robot. By analyzing the data collected by laser radar and usingtabu search algorithm, the above algorithm can identify the optimized sub-goal. Thenthe mobile robot can realize the free-obstacle and optimization path based on fuzzylogic control method by gradual way. Simulation and experiment results show that theproposed algorithm is correct and effective.6. In order to solve the local minimum problem encounted within variousflocking systems in unexpected and irregular obstacle environment, the virtualsub-goal based flocking motion and obstacle avoidance control strategy is proposedby giving individual agents some limited memories. By considering the dynamicanalysis and virtual force decomposition, an overall control algorithm of multi-robotsgroup is devised by transforming the agent-based flocking control strategy. Themulti-robots group can be more effieient to achieve flocking motion in obstacleenvironment.