Traffic Simulation and Autonomous Driving Experiment in VIPLE

Abstract

This paper presents the development of visual programming languages and summarizes the recent research in traffic simulation and autonomous driving experiments in VIPLE (Visual IoT/Robotics Programming Language Environment). The traffic simulator is developed in the Unity programming engine. The traffic simulator allows generation of different traffic patterns, either randomly or guided through a recorded real-world traffic dataset. Traffic experiments can be programmed in VIPLE or in Unity simulator, allowing developers to create such experiments without a deep understanding and lengthy support programming around the environment. Instead, the user can focus on implementing optimal routing algorithms that navigate a vehicle through the city traffic. The autonomous driving environment is based on TORCS simulation environment. A variety of difficulty levels of autonomous driving experiments can be created. Both traffic simulation and autonomous driving experiments can be written in visual programming language VIPLE, as well as in C# and Python within VIPLE environment. VIPLE has been widely used in different courses worldwide. This new advancement in VIPLE can particularly help students to perform traffic control and autonomous driving related programming, machine learning, and artificial intelligence research and experiments.