AUTONOMOUS VEHICLE LOCALIZATION AND MAPPING SYSTEM USING A COMBINATION OF 2D LIDAR AND IMU SENSORS

• Main Author: Satrio Wibowo, Teguh
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/70816
• Institution: Institut Teknologi Bandung

Autonomous vehicles require the development of sensing technology for localization capabilities to guide vehicles in an unknown environment. The ability to recognize and know its location about the environment is one of the important capabilities of an autonomous vehicle when it is in a real environment. A reliable localization system of accurate position and heading information is one of the autonomous vehicle technologies that need to be developed. This thesis proposes a localization estimation and mapping algorithm using Lidar and IMU sensors. The IMU sensor is used to obtain estimated position and heading data with a modification of the double integral algorithm and the Attitude Heading Reference System (AHRS). A denoising filter uses a Butterworth Zero Phase Filter and identification of stationary conditions is carried out to eliminate the effects of noise and drift on the IMU sensor data. The HectorSLAM algorithm is used to get the results of mapping the environment. Sensor data collection is carried out using the Robot Operating System (ROS) platform. The implementation is carried out on a wheeled robot designed with two rear drives using a differential drive system and one wheel in front with a passive system using free wheels. The results of testing the proposed localization algorithm can produce an estimated MSE score position of 0.002204 for the x-axis data and 0.002243 for the y-axis data. The results of the HectorSLAM algorithm test produce an error range of 0.034 to 0.217 meters.