Key Technologies Of Virtual Remote Control Of Quadcopter UAV In Underground Coal Mine

With the further advancement of the construction of smart mines,the automation and intelligence requirements of underground coal mine inspections are increasing.The traditional coal mine underground inspection methods are labor-intensive,and human factors lead to unreliable inspections.The four-rotor Unmanned Aerial Vehicle(UAV)has a simple structure and mature control algorithms.With the help of advanced artificial intelligence technologies such as Machine Vision,Virtual Reality,and Intelligent Control,Intelligent Navigation can be achieved to replace people in the complex environment of coal mines to complete the inspection tasks in key underground mines.The underground environment of coal mines is complex,the flying space of UAV is limited,the difficulty of autonomous flight is high,and the risks are high.Therefore,with the help of technologies such as Virtual Reality,Digital Twins,and Intelligent Control,a virtual remote control technology solution for quadrotor UAV suitable for coal mine roadway inspection is proposed.Construct virtual prototypes of four-rotor UAV and virtual scenes of underground tunnels,study key technologies such as obstacle perception and three-dimensional reconstruction,precise positioning of UAV,and obstacle avoidance navigation to provide efficient cognitive conditions for operators to achieve drone autonomy.The control concept of combining flight and manual remote intervention realizes the "Immersive" remote control of the UAV,which effectively improves the efficiency and safety of the patrol inspection of the four-rotor UAV under the coal mine.Aiming at the problem that the UAV environment is difficult to accurately sense due to underground obstacles and is prone to collision,the intelligent sensing and three-dimensional reconstruction technology of roadway obstacles is studied.Use the binocular vision measurement unit to take pictures of obstacles from different perspectives,and perform stereo matching on the obtained left and right camera images to obtain a parallax map,and then obtain three-dimensional information of the obstacles.Deep fusion of obstacle data and virtual tunnel model through coordinate matching.Use the three-dimensional grid cloning technology to stack the obstacle outline model in the virtual tunnel to realize the local map reconstruction.The virtual prototype driven by flight posture data and the virtual laneway scene dynamically updated in real time provide remote control personnel with control basis for human intervention decision-making.According to the requirements of the quadrotor UAV for stable flight in a complex roadway environment,a four-rotor UAV pose control model is established,and the extended Kalman filter algorithm is used to achieve visual inertial pose data fusion,and coordinate the UAV attitude.The data is coupled to the virtual roadway geometry model to obtain the precise position of the UAV in the roadway.Research the path planning and obstacle avoidance technology of the four-rotor UAV,and propose a compound virtual force field VFF algorithm.The velocity potential field is introduced into the virtual force field obstacle avoidance navigation algorithm to realize the coal mine inspection UAV for unknown or in the roadway.Obstacle avoidance flight of dynamic objects.Aiming at the problem of complicated and unintuitive remote control decision-making caused by the complicated and unintuitive flight data of the four-rotor UAV of the coal mine,the data-driven virtual reality interactive technology of the four-rotor UAV of the coal mine is studied.Using the Unity3D platform to develop a virtual reality interactive platform for a four-rotor UAV in a coal mine,integrating flight data visualization technology,to realize a remote monitoring platform combining three-dimensional viewing and flat-screen display.An optimization scheme for complex scene data loading is proposed,which effectively simplifies complex 3D scene data and ensures the smoothness of scene loading.Developed a virtual-physical data communication interface,feed real-time feedback of quadrotor UAV flight data to virtual control platform,realize remote control of UAV,synchronous control of virtual and real.Build a coal mine four-rotor UAV experimental platform to verify the system functions.Experiments show that the visual virtual simulation system runs stably,and the virtual remote control platform is easy to operate,and can perform obstacle perception and 3D reconstruction,and realize the autonomous obstacle avoidance navigation and inspection control of the UAV.