SLAM Research Based On ROS And Indoor Four Rotor Aircraft

Due to the characteristics of sudden natural disasters, rapid response after disasters is significant to disaster emergency rescue. Abtaining information Timely and fast in disaster plays an important role for rescue strategy. Unmanned aerial vehicles (UAVs) gradually show more and more advantages in disaster rescue.SLAM (simultaneous localization and mapping), also known as CML (Concurrent mapping and localization), concurrently build map and localize. SLAM is first put forward by Smith, Self and Cheeseman in 1988. Due to its important theoretical and application value, it is considered as the key to full autonomous mobile robot by many scholars.This thesis is aimed at SLAM technology with four rotor aircraft, building map of the disaster scene. At the same time, theoretical and technical supports for international air robot competition are prepared. SLAM algorithm on UAV will be conducive to efficient and fast military reconnaissance and rescue and be helpful to people’s life and property security as well.First, the four rotor aircraft system hardware construction is analyzed including the system architecture design, four rotor aircraft rotor aircraft hardware and onboard SLAM sensor.Next, designed software system of rotor aircraft based on APM and ROS including details of APM flight controller, ROS robot operating system and the communication between them is analyzed. And the experiment results show that the APM and ROS on rotor aircraft meet the needs of the deployment of SLAM algorithm.Last, rotor aircraft SLAM algorithm and implementation are analyzed. After describing the general architecture and SLAM classic algorithms, hectorjmapping algorithm is studied and implemented. Finally, the experiment results show that hector_mapping successfully implemented the rotor aircraft SLAM algorithm.