Developing autonomous agricultural tractors demands an integrated approach that combines multibody dynamics, terrain interaction and advanced navigation algorithms. This presentation highlights how physics-based modeling and simulation facilitate the design and validation of vehicle dynamics, perception and decision-making in realistic field conditions using environments like Unreal Engine. We will explore the full autonomy stack, including traversability mapping, path planning, trajectory tracking and sensor fusion with GNSS and INS. By leveraging model-based design, engineers can enhance vehicle performance, improve ride comfort and accelerate autonomous deployment. Attendees will gain insights into bridging the gap between physics-based modeling and real-world autonomy implementation.

• Multidomain physical modeling for simulating tractor dynamics and terrain interactions
• Perception and sensor fusion for real-time environment awareness and obstacle avoidance
• Path planning and control for precision navigation and optimized field coverage
• HIL and SIL testing to validate autonomy algorithms before real-world deployment
• End-to-end workflow from simulation to deployment, enabling scalable and efficient development