Formation Control For Mutiple Automous Underwater Unide Networking

Cooperative control of multiple autonomous underwater vehicles (AUVs) plays an important role on marine scientific investigation and marine development. The formation of multiple AUVs can significantly enhance the application capacities on the marine sampling, imaging, surveillance and communications, etc. According to the spatial distribution of the formation of multiple underwater vehicles, we classify the formation control problem into two formation categories on spatial scales. We consider the large-scale formation control with many AUVs and with long ranges among them and the small scale formation control referring steering angles of AUVs, respectively. The major results are as follows:1. The framework of formation control is investigated and a virtual-leader-follower formation control scheme combinating leader-follower with virtual structure is introduced. Considering the signal attentuation and distortion resulted from underwater communication range constraints, we divide AUVs in space into some clusters according to their relative postions. Leaders of all clusters can track their desired trajectories based on virtual strcture. Followers within all clusters follow their corresponding leader with the leader’s disired tracking trajectory, which realizes the leader-following formation. Numerical simulation results show the effectiveness of the control strategy.2. The design method of finite-time tracking controller is investigated. A compound finite-time tracking control linearization algorithm based on feedback linearization and variable structure is proposed. Firstly, the reaching-control law is used to drive the steering angle trajectory with inertial error in finite time toward the given switching surface neighborhood. When the steering angle trajectory reaches inside the neighborhood, the chattering-free control law based on continuous and smooth state-feedback drives the trajectory to the switching surface precisely while converting the original nonlinear system into a reduced system. Secondy, the state-feedback control law for the reduced system is designed to stabilize the position error in finite time. Numerical simulations show the desired trajectory of AUV is attained fully in finite time without chattering.3. The trajectory-tracking problem for AUV based on nonlinear filtering is considered. Based on the research in the second chapter, the estimation states of the error-system with system noise and measurement noise are used to construct the feedback control law based on extended kalman filter and unscented kalman filter with upper triangular and diagonal matrix decomposition of the error covariance. Numerical simulations show the effectiveness of the control scheme.4. The large-scale formation control of multiple autonomous underwater vehicles based on the consensus algorithms of multi-agent systems with first-order dynamics is investigated. For the consensus problem of AUVs with different virtual leader reference information, we propose two-level finite-time consensus algorithms within a sampling period, which simultaneously carry out the consensus on different reference information and the consensus on different AUVs’states, respectively. Formation control with virtual leader for multiple agents, formation control with single leader based on Leader-follower, hierarchical formation control for multiple agents based on Leader-follower and formation control for multiple agents with constraints on communication ranges are investiaged respectively, and a series of finite-time consensus tracking control protocol for multiple agents are presented.5. The large-scale distributed cooperative control of autonomous underwater vehicles is investigated. We consider the network-induced time-delay and data dropout under large-scale cooperative control in a unified framework. AUVs store received state information from their neighbors and use the historical information to make the consensus when data links has failed or information has been lost based on data sampling and holding. The collicsion-free coordinated control of AUVs is carried out based on finite-time consensus algorithms and artificial potential field. The cooperative control of multiple agents system with variable communication topologies under communication ranges constraints and some agents’dropping-out is analysed with illustrative examples. To suppress the influence of environmental noise and external disturbance, the feedforward and feedback optimal consensus tracking control of multiple agents system based on kalman filtering is studied.6. The small-scale formation control of autonomous underwater vehicles is investigated. Firstly, the finite-time formation control in two-dimension space is considered. The relative positions to the virtual leader are transformed into their desired positions, and the control law in the third chapter is proposed to make the AUVs track the desired trajectories in finite time. Secondly, the finite-time formation control of autonomous underwater vehicles with positions and attitudes in three-dimension space is investigated. A finite-time consensus algorithm for second-order system is proposed, the consensus on velocities of AUV (linear velocity and angular velocity) and positions (displacement and angles) are carried out. We demonstrate the formation control of multiple AUVs with different communication ranges and with constraints on maximum velocity, which show that the finite-time consensus on positions and velocities is obtained.