Webinar | Advanced Applications for Autonomous Off-Road Vehicles Using 4DV-Sim

This webinar features the real-time simulation of two advanced applications for autonomous off-road vehicles, equipped with robots and sensors. 4DV-Sim is used to simulate in real-time (i) off-road vehicles, namely Ackermann-based steering tractors, (ii) sensors such as camera and Lidars, and (iii) serial robotics manipulators. 4DV-Sim runs on Linux 24.04 / Ubuntu. The first application is about an agricultural tractor that circulates among row crops. The virtual tractor is equipped with a 2D camera. Two control algorithms are employed to successfully achieve this maneuver: (i) edge detection when moving within a row and (ii) Dubins path with Stanley controller when U-turning to move from one row to the other. The controller is coded in Python. The co-simulation tool is ROS2 Jazzy. Both Python and ROS2 are running, in real time, on the same computer as 4DV-Sim. The data synchronization mechanism is used to ensure the correctness of the data flow when closing the control loop. The second application is a tractor equipped with a 3-dof serial robotics manipulator that holds a mower at its end to mow grass close to the sidewalk. The tractor is equipped with two Lidars. Two control algorithms are employed to: (i) ensure tracking of the sidewalk and (ii) fold the robotic arm in the presence of obstacles such as trees, rocks, etc. The controller is simulated in MATLAB/Simulink and is running in real time via RT-LAB. The co-simulation tool is Orchestra. Both RT-LAB and Orchestra are running, in real time, on a separate computer from 4DV. The time synchronization mechanism is used to ensure correctness of the data flow when closing the control loop.