Sensor fusion-based approach for the field robot localization on Rovitis 4.0 vineyard robot

• DOI: 10.25165/j.ijabe.20221506.6415
• Publication Date: 2022-12-25T16:45:16Z
• Authors: Jurij Rakun; ; Matteo Pantano; Peter Lepej; Miran Lakota
• Keywords: Odometry; Odometer; Sensor Fusion; Visual Odometry; Robot calibration; Ground truth

This study proposed an approach for robot localization using data from multiple low-cost sensors with two goals in mind, to produce accurate and keep the computation as simple possible. The used wheel odometry, inertial-motion Inertial Motion Unit (IMU), a location fix Real-Time Kinematics Global Positioning System (RTK GPS). Each of is prone errors some situations, resulting inaccurate localization. odometry affected by caused slipping when turning or putting it on slippery ground. IMU produces drifts due vibrations, RTK GPS does not return (semi-) occluded areas. None these enough precise reading sound outdoor environment. To solve this challenge, sensor fusion was implemented prevent possible errors. It worked selecting most readings given moment pose estimation. evaluate approach, different tests were performed, one operating system (ROS) repository other presented Field Robot Localization. first did perform well, while second evaluated comparing orientation estimate ground truth, captured hovering drone above testing ground, which revealed average error 0.005 m±0.220 m estimating position, 0.6°±3.5° orientation. proved that developed field robust be ROVITIS 4.0 vineyard robot. Keywords: localization, IMU, GPS, vineyard, robot, fusion, ROS, precision farming DOI: 10.25165/j.ijabe.20221506.6415 Citation: Rakun J, Pantano M, Lepej P, Lakota M. Sensor fusion-based Rovitis Int J Agric & Biol Eng, 2022; 15(6): 91–95.