Motion Generation Considering Situation with Conditional Generative Adversarial Networks for Throwing Robots
This addresses motion planning challenges for robots in dynamic environments, though it is incremental as it applies existing cGAN methods to a specific robotic task.
The paper tackles the problem of generating appropriate robot motions in cluttered environments where constraints change frequently, by using conditional Generative Adversarial Networks (cGANs) to search a latent space for valid motions, resulting in three times higher accuracy and 2.5 times faster calculation time compared to direct action space search in object-throwing tasks.
When robots work in a cluttered environment, the constraints for motions change frequently and the required action can change even for the same task. However, planning complex motions from direct calculation has the risk of resulting in poor performance local optima. In addition, machine learning approaches often require relearning for novel situations. In this paper, we propose a method of searching appropriate motions by using conditional Generative Adversarial Networks (cGANs), which can generate motions based on the conditions by mimicking training datasets. By training cGANs with various motions for a task, its latent space is fulfilled with the valid motions for the task. The appropriate motions can be found efficiently by searching the latent space of the trained cGANs instead of the motion space, while avoiding poor local optima. We demonstrate that the proposed method successfully works for an object-throwing task to given target positions in both numerical simulation and real-robot experiments. The proposed method resulted in three times higher accuracy with 2.5 times faster calculation time than searching the action space directly.