Experimental Evaluation of Human Motion Prediction Toward Safe and Efficient Human Robot Collaboration

Human motion prediction is non-trivial in modern industrial settings. Accurate prediction of human motion can not only improve efficiency in human-robot collaboration, but also enhance human safety in close proximity to robots. Although many prediction models have been proposed with various parameterization and identification approaches, some fundamental questions remain unclear: what is the necessary parameterization of a prediction model? Is online adaptation of models necessary? Can a prediction model help improve safety and efficiency during human-robot collaboration? These unaddressed questions result from the difficulty of quantitatively evaluating different prediction models in a closed-loop fashion in real human-robot interaction. This paper develops a method to evaluate the closed-loop performance of different prediction models. In particular, we compare models with different parameterizations and models with or without online parameter adaptation. Extensive experiments were conducted on a human-robot collaboration platform. The experimental results demonstrate that human motion prediction significantly enhance the collaboration efficiency and human safety. Adaptable prediction models that are parameterized by neural networks achieve better performance.