Arena-Bench: A Benchmarking Suite for Obstacle Avoidance Approaches in Highly Dynamic Environments

Abstract--The ability to autonomously navigate safely, especially within dynamic environments, is paramount for mobile robotics. In recent years, DRL approaches have shown superior performance in dynamic obstacle avoidance. However, these learning-based approaches are often developed in specially designed simulation environments and are hard to test against conventional planning approaches. Furthermore, the integration and deployment of these approaches into real robotic platforms are not yet completely solved. In this paper, we present Arena-bench, a benchmark suite to train, test, and evaluate navigation planners on different robotic platforms within 3D environments. Existing benchmarks for robot navigation algorithms mostly focus on static environments, but few exist that cover both dynamic I. On that account, OBILE robots are increasingly being employed for various use cases such as last-mile delivery, healthcare we propose Arena-bench, a benchmark suite consisting of services, or operation in hazardous environments [1]. This dynamic environments is essential for the operation of mobile benchmark provides an intuitive interface to design and create robotics. In recent years, Deep Reinforcement Learning (DRL) dynamic scenarios within 2D and 3D simulators based on has accomplished remarkable results for dynamic obstacle Flatland and Gazebo, respectively.