Electric Powertrain Efficiency Improvement for Autonomous Vehicles Using Genetic Algorithms for Optimized Speed Profile Creation

In this paper, a novel method is proposed to improve the electric motor efficiency of autonomous electric vehicles with a single electric motor and single gear (IMIG) by controlling the vehicle’s speed profile during speed changes and maintaining acceptable ride comfort levels by limiting jerk during acceleration and deceleration. The Bezier curve method was used to create smooth speed profiles between different speeds, and a genetic algorithm (GA) optimization approach was used to find the optimal set of parameters for high efficiency, low jerk and high average speed. Comparison with the constant jerk - constant acceleration (CJ-CA) method shows a noticeable improvement in motor efficiency when running an urban drive speed profile. This was achieved by increasing the overall motor efficiency during acceleration and energy recovering during deceleration.