Hybrid Quantum-Inspired Algorithm for Optimizing Fuel Efficiency of Autonomous Vehicle Fleets

Abstract

Fuel Efficiency Optimization for Autonomous Vehicle Fleets is a challenge in intelligent transportation systems with dynamic traffic, different vehicle states, unpredictable road conditions, and cooperative driving limitations. Conventional fleet optimization techniques are less appropriate for large-scale, non-linear, and multi-objective decision problems because they often use deterministic routes or local control rules. A hybrid quantum-inspired method for autonomous fleet fuel-efficiency optimization is presented in this research. This approach consists of four parts: cooperative velocity planning, adaptive route-energy coupling, quantum-inspired population encoding, and local search refinement. Fuel consumption, trip time, vehicle load, distance between vehicles, traffic congestion, and route assignment are all taken into account simultaneously via a fleet-level optimization framework. Mixed-road, urban, and arterial scenarios have all been simulated. When compared to the conventional genetic and particle swarm optimization baselines, the new approach decreased the average fuel consumption by 12.8%, the total fleet energy cost by 9.6%, and the convergence speed by 15.3%. According to the analysis, quantum-inspired search algorithms can enhance the fleet of intelligent vehicles' autonomy and offer strong computational support for environmentally friendly intelligent transportation.