Multi-Objective Reinforcement Learning Framework for Transmission Shift Schedule Optimization

Building upon previous work that successfully employed a Reinforcement Learning (RL) agent for the autonomous optimization of transmission shift programs to enhance fuel efficiency , this paper addresses a critical limitation of that approach: the neglect of human-centric factors. While the prior methodology achieved substantial fuel consumption reductions by training an RL agent in a Software-in-the-Loop (SiL) environment, it did not explicitly account for aspects such as driver comfort and preferences, which are paramount for realworld user acceptance and drivability. This work presents a multi-objective optimization framework extending the artificial calibrator to simultaneously maximize fuel efficiency and enhance driver comfort. The method introduces a modified RL reward function that penalizes undesirable shift behavior to ensure a smooth driving experience (drivability). This new methodology also incorporates a mechanism to capture and integrate driver preferences, moving beyond a purely quantitative fuel-economydriven objective to a holistic, user-focused calibration. Experimental evaluation demonstrates that the extended framework successfully generates a shift strategy that achieves a favorable trade-off between fuel efficiency and drivability, resulting in a more balanced and practical calibration. The ability to integrate these qualitative factors into an automated, data-driven process represents a significant step forward, promising to accelerate the development of powertrain control systems that are both highly efficient and aligned with the expectations of human drivers. This work lays the foundation for future RL-based calibration tools that are capable of addressing the full spectrum of development objectives, from fuel economy to the subtleties of vehicle drivability.