Research On Distributed Coverage Control Of Heterogeneous Unidirectional Mobile Sensor Network

In recent years,wireless mobile sensor networks have developed rapidly.The networks have wide applications in many fields,such as military exploration,environment monitoring and intelligent home and so on.A fundamental topic of the networks is the coverage control problem,where the goal is to optimize the sensors’ positions and achieve the desired coverage performance.Now most of works focus on homogeneous networks.Little attention has been paid to the coverage control problem for heterogeneous sensor networks with different maximum velocties and locomotion constraints.This thesis studies the coverage control problem for a network of heterogeneous mobile sensors with unidirectional locomotion.In general,a coverage cost function is introduced to describe the largest arrival time from the mobile sensor networks to any point on a closed curve.Firstly,the coverage control problem for a network of heterogeneous mobile sensors with unidirectional locomotion on a circle is investigated.Its goal is to globally optimize the sensors’ positions on the circle such that the coverage cost function is minimized.The method of low gain feedback is applied to design a distributed control law for each sensor with first-order dynamics.A sufficient condition is provided to avoid collision between the sensors and an upper bound on the low gain is also given.Then,a necessary and sufficient condition of minimizing the cost coverage function is provided via a partition of the circle.It is proved that the mobile sensors can be driven to the optimal positions such that the coverage cost function is globally minimized by constructing a Lyapunov function.Secondly,since nonuniform roughness can affect the arrival time from the networks to any point,we further study the nonuniform coverage control problem for heterogeneous mobile sensor networks with unidirectional locomotion on a circle.Its goal is to optimize the sensor networks’ positions on the circle with nonuniform roughness such that the coverage cost function is minimized.For a network of mobile sensors with first-order dynamics,the method of low gain feedback is utilized to design a distributed control law,which can avoid collision between the sensors.With the help of the Lyapunov theory,we prove that the sensors can be driven to the optimal configuration on the circle such that the coverage cost function is minimized.Finally,since the environment boundary curve is often non-convex,we further consider the coverage control problem for mobile sensor networks with unidirectional locomotion on a general closed curve.Its goal is to optimize the sensors’ locations on the curve such that the coverage cost function is minimized.The relative distance between sensors is measured using the relative arc-length.A distributed control law for each sensor with first-order dynamics is designed using the method of the low gain feedback,and ensures that the mobile sensors will not collide with each other.Then,a necessary and sufficient condition for the minimization of the coverage cost function is given via a partition of the curve.It is proved that the control law can drive them to the optimal positions on the curve while minimizing the coverage cost function by constructing a Lyapunov function.