Leveraging the Human Operator in the Design and Control of Supernumerary Robotic Limbs

• Published in: IEEE robotics and automation letters 2020-04, Vol.5 (2), p.2177-2184
• Main Authors: Guggenheim, Jacob, Hoffman, Rachel, Song, Hanjun, Asada, H. Harry
• Format: Article
• Language: English
• Subjects: Actuation; Actuators; Algorithms; Control systems; End effectors; Fingers; Grippers; Hand (anatomy); Human augmentation; Human body; Joints (anatomy); Limbs; Muscles; Pelvis; Redundancy; Robot control; superlimbs; supernumerary robotic limbs; Task analysis
• ISSN: 2377-3766; 2377-3766
• DOI: 10.1109/LRA.2020.2970948

A human has over 200 muscles in the body, creating a high degree of flexibility and redundancy in movement. This paper exploits this high degree of redundancy for the actuation and control of Supernumerary Robotic Limbs (SuperLimbs), which are attached to a human body. SuperLimbs containing many active joints tend to be too heavy to wear comfortably. Since SuperLimbs are attached to a human body at their base, Superlimbs can be positioned directly by moving the base with movements of the human body. No active joints are needed for the SuperLimbs in certain directions if the human body can generate the same movements as the SuperLimbs, thus allowing for the design of reduced-actuator Superlimbs. Here, we present a method for quantifying the usable degrees of freedom (DOFs) of a human body for a specific task so that the number of Superlimb actuators can be reduced. The high degree of redundant human DOFs can also be utilized for communication and control. Human's fingers are often redundant for performing a task, e.g. holding a box. Although both hands are busy, some combination of the finger forces is still available for generating signal patterns. An algorithm is developed for generating coded finger force patterns without interfering with the performance of the primary task. Both methods are implemented on a simple SuperLimb and a human subject demonstrates the usefulness of the methods.