This paper describes how RL agents in the Unity Environment can perform parking. The goal of the study is to propose method that makes use of reinforcement learning techniques offered by the Unity ML- Agents framework within Unity’s realistic 3D simulation in order to solve the requirement for autonomous parking solutions. The suggested solution’s design, execution, and assessment are highlighted in the paper. In complex situations, the system offers an adaptive and realistic frame-work for autonomous parking. The outcomes of thorough performance testing and comparative analysis highlight the use fullness and promise of the suggested approach in the area of autonomous car parking. The discussion of the results, difficulties faced, and prospects for additional study and advancement in autonomous car parking technology round up the report.