Energy-Efficient Shortest Path Planning Based on Improved D* Lite Algorithm in Large-Scale Off-Road Environments

Most traditional vehicle path planning algorithms which take the shortest path length as the optimization objective may not be able to generate safe and energy-efficient global paths in complex off-road environments. To address this problem, an improved D
^*
Lite algorithm that comprehensively considers the impact of distance and energy cost on the mobility of unmanned ground vehicles (UGVs) is proposed in this paper. After finding a physical model of energy consumption including distance, slope, and friction coefficient, the off-road environment is modeled as a network with nodes and weighted edges so that the D
^*
Lite algorithm can be applied to find the path with the minimum edge costs. Especially, the data storage structure, tie-breaking criteria, and retrieval strategy of D
^*
Lite are optimized to reduce search time and improve planning efficiency. Simple additive weighting is used to solve the multi-objective optimization problem and users can set different weights for optimization variables according to their preferences. The simulation results revealed that the improved D
^*
Lite algorithm generated a more feasible, energy-efficient, and safer route with higher computational efficiency.