Research On UAV-based Inspection Method For Oil Tanker Cargo Hold Structure

At present,hull structure inspection mainly relies on facilities such as scaffolding,aerial vehicles and portable ladders,and these methods have problems such as low efficiency,high cost and high danger.The use of drones for hull inspection,real-time transmission of highdefinition images to the ground monitoring end,can eliminate the inconvenience and hidden dangers associated with the construction of facilities,can greatly reduce inspection costs,improve inspection efficiency and reduce the risk of high-altitude operations.Nowadays,the development of micro UAVs is gradually accelerating,and widely used.The research on autonomous navigation,autonomous obstacle avoidance and other related aspects for UAVs has become a research hotspot around the world.Uavs are typically used in open outdoor environments,at high altitudes,or in well-lit indoor environments.While the UAV in the case of poor GPS signal and poor lighting,GPS and visual navigation system may fail.For example,the use of UAVs for ship hull inspection.In the steel hull structure,due to signal shielding,light source instability and other reasons,the traditional means can not make UAVs achieve stable flight,and it is also difficult to complete the autonomous obstacle avoidance.This paper focuses on the quadrotor UAV positioning and global path planning algorithm based mainly on sensors such as Li DAR from the actual environment of ship inspection without GPS and low light environment.And the quadrotor UAV positioning and navigation algorithm is optimized according to the characteristics of the bulk carrier cargo environment.The research in this paper will be carried out in the following aspects:First of all,the quadrotor UAV autonomous navigation system is designed according to the existing mature autopilot technology.And a set of UAV autonomous navigation platform architecture is built,and the UAV simulation platform is configured based on the idea of the three-layer system layout of physical layer,protocol layer and application software layer in the ROS system.The system is divided into flight control,data transmission and obstacle avoidance navigation and other related modules,and the software level design and implementation of the composition of each module.Select the appropriate sensors and equipment to build the UAV capable of autonomous navigation and to check the usability of the hardware devices.Then build the UAV autonomous cruise system in the virtual in-the-loop simulation environment.Secondly,the map form for navigation and the way the UAV is positioned indoors are investigated.Occupancy grid map are selected to construct the flight environment model of the UAV.Comparing multiple map building algorithms,the optimized Cartographer algorithm is used for UAV’s positional estimation,and the optimized algorithm reduces the effect of Li DAR frequency fluctuations on map details.Subsequently,to address the problem of slow convergence in the application of traditional path planning algorithms,the optimized RRT* algorithm is used to realize obstacle avoidance and path planning of UAVs in a typical ship cabin environment by combining the actual situation in the ship cargo.Finally,the internal structure of the cargo hold of the bulk carrier is built in the simulation environment to test the effectiveness of the algorithm,and the functions of the UAV system proposed in this paper are simulated and tested one by one in the PC.The test results show that the system can complete 2D automatic obstacle avoidance and automatic flight at various heights in the cargo environment according to the preset targets,and can adjust its own position to complete the ship inspection.