Porting Evaluation Framework for Self-Driving Software Based on ROS 2

Self-driving systems are being actively researched to increase the popularity of autonomous vehicles. In recent years, Robot Operating System (ROS) 2, is a middleware that provides libraries and tools for robot development, and it has been used for self-driving systems. However, a method for testing the safety of self-driving systems has not yet been established. In addition, when porting self-driving software, testing is important to confirm that the self-driving software works properly. This paper proposes a porting evaluation framework for self-driving software based on ROS 2. The proposed framework stores and compares the output results of modules that are working accurately and previously ported modules with the same functionality. Experimental results show that the proposed framework can evaluate the porting of self-driving software by comparing the output values of the modules.

Self-driving systems are being actively researched to increase the popularity of autonomous vehicles. In recent years, Robot Operating System (ROS) 2, is a middleware that provides libraries and tools for robot development, and it has been used for self-driving systems. However, a method for testing the safety of self-driving systems has not yet been established. In addition, when porting self-driving software, testing is important to confirm that the self-driving software works properly. This paper proposes a porting evaluation framework for self-driving software based on ROS 2. The proposed framework stores and compares the output results of modules that are working accurately and previously ported modules with the same functionality. Experimental results show that the proposed framework can evaluate the porting of self-driving software by comparing the output values of the modules.