TC AgRA – Webinar 39 – Improving Autonomous Orchard Vehicles Trajectory TrackingBy Robolab Technologies In Robotics
Dr. Gokhan Bayar is a professor of Mechanical Engineering Department at the Bulent Ecevit University in Zonguldak, Turkey, where he leads projects and teaches courses in robotics, mechatronics, and control systems. His main interests are the design, development, modeling, and control of mobile robots and autonomous ground vehicles; and low-cost and high-resolution lidar systems for autonomous vehicle applications—especially as applied to problems in agriculture. Dr. Bayar earned M.Sc., and Ph.D. degrees in Mechanical Engineering Department from the Middle East Technical University, Ankara, Turkey. During his Ph.D. studies, he spent a year in the Field Robotics Center, Robotics Institute, Carnegie Mellon University, in Pittsburgh, where he developed a significant part of the theory and conducted many of the experiments presented today.
The main goal of this work is to improve trajectory tracking by an autonomous orchard vehicle by compensating for the effects of slippage. First, a wheel slippage estimator is developed and included into the system’s dynamic model. Then, appropriate back-stepping controllers are used to control the steering and velocity of the vehicle. A high accuracy positioning system is used to estimate the slip velocities of the rear wheels in the longitudinal and lateral directions. In addition to the slip compensation method, this talk will present our work on vehicle localization in the orchard using only a front-facing laser rangefinder and a new method for row turning. The methods have been implemented and tested in practice on a four-wheeled orchard vehicle with steerable front wheels and an electric rear motor in snow- and mud-covered terrain. The experimental results show that the trajectory tracking performance of the autonomous orchard vehicle improves when slippage is considered in the control loop.