ALPHA- α and Bi-ACT Are All You Need: Importance of Position and Force Information/ Control for Imitation Learning of Unimanual and Bimanual Robotic Manipulation With Low-Cost System

Autonomous manipulation in everyday tasks requires flexible action generation to handle complex, diverse real-world environments, such as objects with varying hardness and softness. Imitation Learning (IL) enables robots to learn complex tasks from expert demonstrations. However, a lot of existing methods rely on position/unilateral control, leaving challenges in tasks that require force information/control, like carefully grasping fragile or varying-hardness objects. As the need for diverse controls increases, there are demand for low-cost bimanual robots that consider various motor inputs. To address these challenges, we introduce Bilateral Control-Based Imitation Learning via Action Chunking with Transformers(Bi-ACT) and“A” “L”ow-cost “P”hysical “Ha”rdware Considering Diverse Motor Control Modes for Research in Everyday Bimanual Robotic Manipulation (ALPHA- $\alpha $ ). Bi-ACT leverages bilateral control to utilize both position and force information, enhancing the robot’s adaptability to object characteristics such as hardness and shape. The concept of ALPHA- $\alpha $ is affordability, ease of use, repairability, ease of assembly, and diverse control modes such as position, velocity, and torque mode. In our experiments, we conducted detailed analysis of Bi-ACT in unimanual tasks involving objects with varying hardness, shape, and weight, confirming its superior performance and adaptability. We also applied Bi-ACT to bimanual tasks such as “Egg Handling” and “Open Cap” using ALPHA- $\alpha $ . The experimental outcomes demonstrated a high success rate in bimanual operations, validating the effectiveness of our approach in real-world scenarios. These results suggest that Bi-ACT and ALPHA- $\alpha $ can advance automation in daily life and industrial settings. Video available at: https://mertcookimg.github.io/alpha-biact/