Emergence of Different Modes of Tool Use in a Reaching and Dragging Task
This work provides insight into the spontaneous emergence of complex tool-use strategies for researchers studying embodied intelligence and reinforcement learning.
This paper explores the emergence of diverse tool-use behaviors in a simulated reaching and dragging task where a robotic arm must use a tool to move an object to a target. The deep reinforcement learning controller developed a rich repertoire of unexpected strategies, including hitting, error correction, and throwing, in addition to expected dragging.
Tool use is an important milestone in the evolution of intelligence. In this paper, we investigate different modes of tool use that emerge in a reaching and dragging task. In this task, a jointed arm with a gripper must grab a tool (T, I, or L-shaped) and drag an object down to the target location (the bottom of the arena). The simulated environment had real physics such as gravity and friction. We trained a deep-reinforcement learning based controller (with raw visual and proprioceptive input) with minimal reward shaping information to tackle this task. We observed the emergence of a wide range of unexpected behaviors, not directly encoded in the motor primitives or reward functions. Examples include hitting the object to the target location, correcting error of initial contact, throwing the tool toward the object, as well as normal expected behavior such as wide sweep. Also, we further analyzed these behaviors based on the type of tool and the initial position of the target object. Our results show a rich repertoire of behaviors, beyond the basic built-in mechanisms of the deep reinforcement learning method we used.