Future Predictive Success-or-Failure Classification for Long-Horizon Robotic Tasks
This addresses the problem of automating condition design for robotic arm tasks, but it appears incremental as it builds on existing planning and prediction methods.
The paper tackles the time-consuming and manual design of conditions for optimization-based action planning in long-horizon robotic tasks by proposing a future-predictive success-or-failure classification method, which uses an end-to-end approach and a regularization term to improve performance, as demonstrated in experiments.
Automating long-horizon tasks with a robotic arm has been a central research topic in robotics. Optimization-based action planning is an efficient approach for creating an action plan to complete a given task. Construction of a reliable planning method requires a design process of conditions, e.g., to avoid collision between objects. The design process, however, has two critical issues: 1) iterative trials--the design process is time-consuming due to the trial-and-error process of modifying conditions, and 2) manual redesign--it is difficult to cover all the necessary conditions manually. To tackle these issues, this paper proposes a future-predictive success-or-failure-classification method to obtain conditions automatically. The key idea behind the proposed method is an end-to-end approach for determining whether the action plan can complete a given task instead of manually redesigning the conditions. The proposed method uses a long-horizon future-prediction method to enable success-or-failure classification without the execution of an action plan. This paper also proposes a regularization term called transition consistency regularization to provide easy-to-predict feature distribution. The regularization term improves future prediction and classification performance. The effectiveness of our method is demonstrated through classification and robotic-manipulation experiments.