Dynamic Adjustment of Reward Function for Proximal Policy Optimization with Imitation Learning: Application to Automated Parking Systems | IEEE Conference Publication | IEEE Xplore