Learning Diverse Skills for Local Navigation under Multi-constraint Optimality
This work addresses the problem of generating diverse and agile behaviors for robotic navigation, which is incremental as it builds on prior quality-diversity trade-off methods.
The paper tackled the challenge of achieving diverse behaviors in robotics without compromising task performance by framing it as a constrained optimization problem, demonstrating on a quadruped robot navigation task that diverse policies can be obtained while meeting multiple reward constraints, with successful real-world transfer to a 12-DoF robot.
Despite many successful applications of data-driven control in robotics, extracting meaningful diverse behaviors remains a challenge. Typically, task performance needs to be compromised in order to achieve diversity. In many scenarios, task requirements are specified as a multitude of reward terms, each requiring a different trade-off. In this work, we take a constrained optimization viewpoint on the quality-diversity trade-off and show that we can obtain diverse policies while imposing constraints on their value functions which are defined through distinct rewards. In line with previous work, further control of the diversity level can be achieved through an attract-repel reward term motivated by the Van der Waals force. We demonstrate the effectiveness of our method on a local navigation task where a quadruped robot needs to reach the target within a finite horizon. Finally, our trained policies transfer well to the real 12-DoF quadruped robot, Solo12, and exhibit diverse agile behaviors with successful obstacle traversal.