Research Of Multi-UAV Load-Balanced Task Planning Algorithm

In UAV-WSN system, multiple UAV work together to perform a task set is an efficient way. The total travel distance and task assignment have an impact on the overall performance. In the current, the research of UAV-WSN is mainly focused on UAV or multi-UAV path planning, collision avoidance between UAVs and obstacl es, as well as attacks to specific targets and other issues. The load balancing problem across multiple UAV mission is fresh. This article is based on an improved k-means clustering algorithm, in the case of the overall mission is in certain, to make the assignment of load as balancedas possible for multi-UAV collaborative environment. Simulation results show that the performance of the load balancing algorithm is well achieved and the optimum task planning scheme is obtained. This improved algorithm has certain value in theoretical and practical.Since the last world 1990 s, UAV(Unmanned Aerial Vehicle) also received attention by many countries in the lead of American and gained rapid development. Compared with traditional manned aircraft, UAVs with a small er, flexible, cheap, sustainable life and other characteristicscan better adapt to the military field investigation, environmental monitoring, search and rescue personnel.Since the mobility and sustainable life, therefore, the UAV used in WSN can not only simplify network design, but also can reduce energy due to the communication between the sensor nodes caused by consumption, thereby extending the lifetime of the network.With the mission grew, task types and complexity increases or other factors, based on a single UAV cannot be timely and effective completion of the task, Multi-UAV began to get public attention. In Multi-UAV system, single UAV can collaborative work and able to complete mission together. On the other hand, due to the presence of multiple UAV system, UAVs can collect information in parallel manners and UAV parallelism may result in a collision between each other, and the task planning between UAVs is more complex.In previous studies, the main concern of authors is path planning of UAV or UAVs, how to avoid the collision with obstacles, attack on specific targets and other issues, but load balancing between multiple UAVsrarely considered. In this paper, the main consideration is how to collaborate accomplish specific tasks across multiple UAV efficiently. The tasks allocation method of multiple and the overall flight distance play a vital role for improving the performance of system. Based on the k-means clustering algorithm, an improved algorithm is proposed. Under the amount of overall mission is certain, try to make load distribution between UAVs as balance as possible. During the experiment, we use Matlab as a simulation tool for modeling the problem, use this new algorithm to divide mission, compare the result with k-means algorithm and based on comparison results to analyze the performance of the new algorithm.