Optimal Path Planning for Wheel Loader Automation Enabled by Efficient Soil-Tool Interaction Modeling
By: Armin Abdolmohammadi , Navid Mojahed , Bahram Ravani and more
Potential Business Impact:
Saves fuel by making digging machines smarter.
Earthmoving operations with wheel loaders require substantial power and incur high operational costs. This work presents an efficient automation framework based on the Fundamental Earthmoving Equation (FEE) for soil-tool interaction modeling. A reduced-order multi-step parameter estimation method guided by Sobol's global sensitivity analysis is deployed for accurate, online excavation force prediction. An optimal control problem is then formulated to compute energy-efficient bucket trajectories using soil parameters identified in the previous digging cycle. High-fidelity simulations in Algoryx Dynamics confirm accurate force prediction and demonstrate 15-40% energy savings compared to standard paths. The total computation time is comparable to a single digging cycle, highlighting the framework's potential for real-time, energy-optimized wheel loader automation.
Similar Papers
Optimizing wheel loader performance -- an end-to-end approach
Computational Engineering, Finance, and Science
Makes diggers load dirt faster and smarter.
In-Situ Soil-Property Estimation and Bayesian Mapping with a Simulated Compact Track Loader
Robotics
Helps robots dig and move dirt anywhere.
Calibrated Dynamic Modeling for Force and Payload Estimation in Hydraulic Machinery
Robotics
Helps diggers know how much dirt they're lifting.