How to select and use tools? : Active Perception of Target Objects Using Multimodal Deep Learning
This addresses the challenge of robot adaptability in tool use for domestic applications, though it is incremental as it builds on prior work with limited adaptability.
The paper tackles the problem of enabling robots to select and use tools for domestic tasks by developing a multimodal deep learning model that learns tool-object-action relations from sensorimotor data, resulting in the robot successfully transferring ingredients with tools even for unknown targets.
Selection of appropriate tools and use of them when performing daily tasks is a critical function for introducing robots for domestic applications. In previous studies, however, adaptability to target objects was limited, making it difficult to accordingly change tools and adjust actions. To manipulate various objects with tools, robots must both understand tool functions and recognize object characteristics to discern a tool-object-action relation. We focus on active perception using multimodal sensorimotor data while a robot interacts with objects, and allow the robot to recognize their extrinsic and intrinsic characteristics. We construct a deep neural networks (DNN) model that learns to recognize object characteristics, acquires tool-object-action relations, and generates motions for tool selection and handling. As an example tool-use situation, the robot performs an ingredients transfer task, using a turner or ladle to transfer an ingredient from a pot to a bowl. The results confirm that the robot recognizes object characteristics and servings even when the target ingredients are unknown. We also examine the contributions of images, force, and tactile data and show that learning a variety of multimodal information results in rich perception for tool use.