Research On Semi-flocking Control Strategy In Mobile Sensor Networks

Mobile sensor networks(MSNs)are highly flexible and scalable,which can be extensively used in wide-area surveillance,search-rescue missions,and pursuit-evasion operations.These typical application scenarios require MSNs with fast and efficient widearea coverage and multi-target tracking capabilities.Meanwhile,there are requirements for mobile nodes with operational flexibility,task processing efficiency,environmental adaptability,and collaboration capabilities.Due to many constraints and uncertainties in real-world applications,such as complex operating environment,changing network topology,and inefficient information exchange,coordinations among mobile nodes are seriously affected,thus leading to system instability and unsatisfactory monitoring capability.Considering the increasing performance requirements and diverse environment constrains,it is challenging to design a suitable and efficient coordination control scheme for MSNs to complete the multi-target tasks.Based on the theories of consensus control,artificial potential field and flocking control,this thesis proposes a distributed semi-flocking algorithm that can rapidly cover a specified area and track multiple targets.Then,a path planning approach is incorporated into the semi-flocking control for further reducing target detection time in fixed rough terrains.Finally,considering area coverage,target tracking and energy-saving performance simultaneously,this thesis proposes an improved semi-flocking algorithm for MSNs operating in unknown complex terrains,further improving the multi-target tracking capability and reducing the energy consumption due to the movements of mobile nodes.The proposed semi-flocking algorithm can manage MSNs by using simple rules and local communication strategies.Mobile nodes can adaptively switch their operation modes according to the task requirements to maximize their area coverage and minimize their target tracking time.This work can provide valuable technical reserve and theoretical support for the efficient operation of MSNs in real-world monitoring scenarios.The main research contents and goals of this thesis are as follows:First,in order to enable MSNs to achieve desired area coverage and target tracking capabilities,this thesis proposes a distributed semi-flocking algorithm based on local communication strategy.The topologies of the network of MSNs are switching rapidly due to the movement of mobile nodes,meanwhile the complex operating environment will affect the quality and stability of the communication links among mobile nodes.To address the problem of inefficient information exchange in MSNs,this thesis designs an efficient local communication strategy for mobile nodes.Based on information exchanges and interactions among mobile nodes,they can switch between the searching mode and the tracking mode according to different task requirements and environmental factors.Furthermore,this thesis provides a mode switching mechanism for combining the advantages of the flocking algorithms and the anti-flocking algorithms.Under the control of the mode switching mechanism,some mobile nodes are capable of forming small formations around each target,and the remaining nodes continue to be responsible for maximizing their area coverage.In order to enhance the target tracking ability of mobile nodes and improve the operational stability of the system,this thesis considers the optimization of the input control of mobile nodes.In addition,for tracking multiple targets with different priorities,this thesis proposes a new target selection method,which can assign mobile nodes to each target according to the priorities of the targets.Second,considering that multiple terrain constraints will affect the motion performance of mobile nodes,MSNs operating on rough terrains have problems with lowefficiency target tracking.In order to reduce the target tracking time of MSNs operating in fixed rough terrains,a path-planning-enabled semi-flocking algorithm is designed.For fixed rough terrains,information about the terrains(such as terrain fluctuations,vegetation density,and soil characteristics)can be obtained.Based on these detailed terrain information and the characteristics of mobile nodes,a reasonable model can be constructed to represent the motion speed limits of mobile nodes in different sub-regions.After modeling the fixed rugged terrain,this thesis designs a high-efficiency heuristic path planning algorithm,which can find the fastest path for mobile node to approach their target.Under the control of the path-planning-enabled semi-flocking algorithm,mobile nodes can not only cooperate with each other to search for a specified area,but also quickly discover and track all the detected targets.The proposed algorithm improves the operational efficiency and adaptability of MSNs on fixed rough terrains.Third,for unknown complex scenarios that are difficult to construct terrain models,such as polar regions,remote islands,and enemy battlefields,it is necessary to design more suitable and efficient algorithms for MSNs to improve their monitoring efficiency.This thesis proposes a Temnothorax Albipennis(T.albippennis)migration inspired semiflocking algorithm to improve the multi-target tracking performance of MSNs on unknown complex terrains.In the proposed algorithm,an efficient coordination mechanism in the T.albippennis migration model is introduced into the semi-flocking algorithm.Mobile nodes can more effectively exchange information and coordinate with their neighboring nodes,and can track all targets in the specified area rapidly.On the other hand,considering that in most MSNs applications,the motion of mobile nodes consumes most of their energy.However,the energy storage of mobile nodes is limited,thus,it is important to reduce the energy consumption of mobile nodes in such scenarios.In order to reduce the energy consumption of mobile nodes in rough terrain while ensuring their efficient area coverage and target tracking capabilities,a terrain adaptive force and a new navigation goal selection method are integrated into the proposed algorithm.The proposed method can maneuver mobile nodes to move along the energy-saving path to achieve excellent area coverage and target tracking performances with lower energy consumption.