Reinforcement learning in car control: A brief survey

Autonomous Vehicles (AVs) from a far foggy dream and cartoon Action, are becoming more and more real nowadays. Realizing this dream can have many benefits both in terms of improvement of the traffic flow as well as reduction of pollution and greenhouse gases. Researchers are continuously developing existing technologies and creating new solutions to achieve stable and safe control algorithms for AVs. Encouraged by their work, the author wanted to make his own contribution to build a better transportation system and taking care of the environment. This paper is a first step - a summary of currently published ideas on the application of reinforcement learning algorithms to the problem of AVs. The paper breaks down the tasks to their nature and level of difficulty, also shows various models of the environment, actions and reinforcement signals outlined at just under 50 papers. Additionally, the number of potential development areas has been highlighted.