Autonomous Moving Target Tracking In Cluttered Outdoor Environments With A Micro Aerial Vehicle

In recent years,Micro Aerial Vehicle(MAV)becomes more and more popular in human production and life,but most aircrafts at this stage can only fly in open environments with human remote control.How to let the MAV equipped with different sensors and algorithms have the ability to track moving target autonomously in cluttered outdoor environments has become a hotspot in the field of mobile robots in recent years.This paper proposes a MAV autonomous moving target tracking method for cluttered outdoor environments.Equipped with a micro onboard computer,visual sensors,a 2D laser radar and other sensors,the MAV can fly with no collision in the process of moving target tracking.The basis of the MAV autonomous moving target tracking is to find an effective and accurate target tracking algorithm.This paper has compared several visual target tracking algorithms,and finally proposed and chosen TLD-KCF algorithm for target tracking and position estimation to cooperate with MAV's autonomous path planning.In addition,this paper proposes a robot path planning algorithm based on elliptical tangent structure.In this algorithm,circumscribed ellipses are developed to surround relevant obstacle regions for dichotomous search of paths.Besides,a series of processes such as elliptical tangent generation and elliptical common tangent generation is obtained to get all alternative paths,and finally select the best path by applying cost function.Unlike traditional algorithms,this one can overcome problems such as a large computation,path is not smooth enough and so on,which can meet the requirements of MAV's fast path planning.In this paper,a M100 aircraft designed by DJI is used as the platform to verify the practicability and effectiveness of the MAV's autonomous moving target tracking system in cluttered outdoor environments.