cRoK: A Composable Robotics Benchmark

Selecting an optimal robot and configuring it for a given task is currently mostly done by human expertise or trial and error. To evaluate automatic selection and adaptation of robots to specific tasks, we introduce a benchmark suite encompassing a common format for robots, environments, and task descriptions. Our benchmark suite is especially useful for modular robots, where the creation of the robots themselves creates a host of additional parameters to optimize. The benchmark defines this optimization and facilitates the comparison of solution algorithms. All benchmarks are accessible through a website to conveniently share, reference, and compare solutions.