Abstract
This paper introduces a decentralised motion planning method for Autonomous Haulage Systems (AHS) using Prioritised Multi-Agent Model Predictive Path Integral Control (P-MA-MPPI). Our approach incorporates vehicle prioritisation, addressing a challenge that has previously limited decentralised MPPI frameworks, to enhance navigation safety and efficiency by exchanging future trajectories among agents. The experimental findings indicate a reduction in collision and deadlock rates when compared to an existing MPPI-based approach, alongside about a 20% increase in energy efficiency in complicated scenarios where three vehicles simultaneously enter an intersection. This research contributes to scalable and sustainable operation strategies, improving the efficiency and the environmental impact of future mine automation.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
