Mutual Awareness And Mission Collaboration Of Multiple Aerial Robotics

Aerial Robot has been widely concerned as the new hot-spot technology. The small-size aerial robots’ navigation in GPS-denial environment is one of the big issues in this area, yet there are plenty of new navigation method was proposed. Visual Navigation is a significant one among them. This paper, whose experimental platform is the quadrotor aircraft, mainly discusses the method of mutual awareness amongst multiple aerial robotics, and has done some experiments of mission collaboration.Based on fiducial marker system called ArUco, the posture recognition and localization problem is studied. By arranging several ArUco visual identifications in the limited circumstance, images with ArUco markers are took by the aerial robot onboard cameras, and used for detecting and extracting the visual identifications to obtain the corner points of the markers. The PnP problem is solved through the relationship between the known world coordinates and image coordinates of the corner points, thus can get the posture and position of the aerial robot.The simplified mathematical model of AR.Drone aircraft is established. An evaluation function is construct to achieve the optimum path planning from a point to a line. The process of flying the aircraft from one point into a line tangently is decomposed into time intervals, in each interval there is only one controlled quantity functioned, an evaluation function is used to determine the optimal controlled quantity to the current time interval, so as to search for an optimal control sequence, under the action of the control sequence, aircraft can gently cut in straight trajectory.The software framework of mutual awareness and mission collaboration system for multiple Aerial Robotics is designed on ROS. An information center module is built to achieve the mutual awareness. At last a simple experiment has been carried on in order to test the capability of mission collaboration for multiple aerial robots. The experimental results show that the system has good coordination ability, to verify the feasibility and accuracy of this algorithm.