Formation Control And Virtual Simulation Of Multiple Autonomous Underwater Robots

Autonomous Underwater Vehicle (AUV) has the advantages of small volume, lightweight, low cost, strong autonomy etc., which has been widely used in ocean exploration,target search and rescue, etc. Because AUV control system is the key factor for itsmaneuverability, so the research of AUV control system, design of algorithm and simulationhas the important practical value and engineering significance. Compared with a single AUV,multi-AUV can undertake more complex multi-objective collaborative tasks. Accordingly, themulti-AUV control system not only has a single AUV functions, but also need to havecoordination control ability among AUVs. Currently, multi-AUV coordination control systembecomes one of the main research directions in AUV control study.Multi-AUV control problem is studied in this paper. From the perspective of the controlalgorithm and visual simulation, AUV formation coordination control, path planning andobstacle avoidance are studied to improve the dynamic quality of multi-AUV coordinationcontrol and verify the Engineering applicability of the proposed algorithm with a moreintuitive form. Major works of this paper are as follows:Part I: formation coordination control. The Multi-AUV system is the leader-followerstructure. Formation error model is derived from AUV kinematic model with distance andattitude among AUVs which is from formation control. Lyapunov function is selected byLyapunov stability theory. And then the design algorithm of follower AUV controller isachieved.Part II: path planning. The ant colony algorithm is used for the leader in multi-AUVsystem to plan path. Octree is used to compress environmental information and build theunderwater environment model. The improved ant colony algorithm is used for path planning.The integrated heuristic information is constructed by the distance between the robot andobstacles and distance between robot and goal. The penalty factor is introduced and applied tothe path found by ant in order to eliminate the path which is near obstacles. As a result, thepath can keep a safe distance to obstacles and is the shortest.Part III: obstacle avoidance. The artificial potential field method is used to increaseobstacle avoidance capability for the follower. The traditional artificial potential field methodis improved, which the distance between the robot and target is added in the repulsivepotential field function to avoid potential field failure when the target is close to the obstacles.Formation and obstacle avoidance algorithm is integrated. When distance between follower and obstacle is less than a particular threshold, the follower uses the improved artificialpotential field method for obstacle avoidance. On the contrary, it uses formation controlalgorithm for formation.Part IV:3D visual simulation. OpenGL and3ds Max are used to build underwater virtualenvironment. Firstly, Three-dimensional models of obstacles and AUV are built by3ds Maxand export ASE model file. Secondly, OpenGL is used to establish underwater scene andrender the AUV and obstacles model. The multi-threading technology is used to simulateAUV sport of the real environment. Java language is used to achieve the path planningalgorithm and integrated algorithm of formation control and obstacle avoidance. And OpenGLdisplays the AUV state in virtual environment achieve formation visualization.