主催: 一般社団法人 日本機械学会
会議名: ロボティクス・メカトロニクス 講演会2017
開催日: 2017/05/10 - 2017/05/13
In recent years, several robot simulators and open source game engines have been developed. Open source game engines such as Unity and Unreal Engine 4 help to create the simulation environment. Therefore, we implemented a simulator environment by connecting a game engine and a robot middleware. The requirements for the simulator are follows: (1) To create a simulation environment with the high degree of freedom. (2) To support the multiple OS. In this study, we selected Unity and OpenRTM-aist. To assess the machine learning performance, the training data can be collected and the learned model can be verified in the simulator environment. As an example of the implementation of the simulation environment using this framework, we developed a simulator for autonomous driving systems.