This paper presents an obstacle-aware feed pusher robot that uses multimodal sensor fusion and an embedded control system to navigate a dairy farm environment safely while performing automated feed pushing. The proposed robot integrates multiple sensors – including ultrasonic rangefinders for obstacle detection and optical line-following sensors for navigation – to perceive its surroundings. An embedded microcontroller processes the fused sensor data in real-time to control the robot’s movements, enabling both autonomous operation and manual override via remote control. We detail the robot hardware architecture and control algorithms, including a mode-switching mechanism for safe obstacle avoidance. The system was implemented on a prototype feed pusher and evaluated in a simulated dairy farm setting. Results demonstrate the robot can effectively follow a predetermined path along the feed alley, push the rejected feed back, and automatically detect and stop for unexpected obstacles such as animals, humans or objects in its path. This solution aims to reduce labor requirements and increase feeding consistency, contributing to improved dairy farm management. The paper also discusses design considerations, current limitations, and future enhancements – including compliance with safety standards, digital twin simulations, and Internet of Things connectivity – to further integrate the robot into precision livestock farming systems.
Multimodal sensor fusion and embedded control for obstacle-aware feed pusher robotics in precision livestock farming
Fourth International Conference on Digital Technologies, Optics, and Materials Science (DTIEE 2025) ; 2025 ; Bukhara, Uzbekistan
Proc. SPIE ; 13662 ; 1366213
05.06.2025
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Springer Verlag | 2024
|Engineering Index Backfile | 1938