Adaptive Anthropomorphic Robots

Adaptive Neuromorphic Prosthetics represent a groundbreaking fusion of neuroscience and artificial intelligence, leveraging advanced algorithms and methods developed within the realm of neuromorphic systems. By emulating the brain's neural networks, these prosthetics achieve real-time responsiveness, energy efficiency, and adaptability to various user needs and environmental conditions. We utilize spiking neural networks and other neuromorphic computing techniques to enable more natural and intuitive control of prosthetic limbs. Our approach enhances the user's experience by providing smoother movements, quicker adjustments, and improved integration with biological systems. We focus on non-invasive technologies that directly interface the environment with the residual arm or legs, allowing for a more seamless and effective user experience.

Features:

- Name: AIzuHand I

- Total Weight: 422g (276g without controller)

- Control: sEMG

- DoF: 5

- Feedback: No

- Mode: AN/SN

Features:

- Name: AIzuHand H

- Weight: 492g

- Control: sEMG, RM

- DoF: 7

- Feedback: No

- Mode: AN/SN

Features:

- Name: AIzuHand II

- Weight: --

- Control: sEMG

- DoF: 7

- Feedback: Yes (Temperature)

- Mode: AN

Cheng Hong, Sinchhean Phea, Khanh N. Dang, Abderazek Ben Abdallah, ''The AIzuHand Neuromorphic Prosthetic Hand,'' ETLTC2023, January 24-27, 2023.
A myoelectric prosthetic allows manipulation of hand movements through surface electromyogram (EMG) signal generated during muscle contraction. However, the lack of an intuitive personal interaction interface leads to unreliable manipulation performance. For better manipulation, sEMG-based prosthetic systems require the user to understand their EMG signal levels and adjust the thresholds for better movements. A practical prosthetic is still being challenged by various factors inherent to the user, such as variation in muscle contraction forces, limb positional variations, and sensor (electrode) placements. Thus, a personal interface that can adjust the prosthetic parameters across amputees is required. In addition, this interface is also needed for some targeted users in pediatrics who require external assistance for control. In this work, we present a low-cost real-time neuromorphic prosthesis hand, AIzuHanda, with sensory motor integration. We aim to develop solutions for controlling prosthetic limbs to restore movement to people with neurologic impairment and amputation. The AIzuHand system is empowered by a user-friendly mobile interface (UFI) for calibration, real-time feedback, and other functional tasks. The interface has visualization modules, such as an EMG signal map (EMG-SM) and mode selection (MS), for adjusting the parameters of the prosthetic hand. Furthermore, the interface helps the user establish a simple interaction with the prosthetic hand.

Mark Ogbodo, Abderazek Ben Abdallah, ''Study of a Multi-modal Neurorobotic Prosthetic Arm Control System based on Recurrent Spiking Neural Network,'' ETLTC2022, January 25-28, 2022
The use of robotic arms in various fields of human endeavor has increased over the years, and with recent advancements in artificial intelligence enabled by deep learning, they are increasingly being employed in medical applications like assistive robots for paralyzed patients with neurological disorders, welfare robots for the elderly, and prosthesis for amputees. However, robot arms tailored towards such applications are resource-constrained. As a result, deep learning with conventional artificial neural network (ANN) which is often run on GPU with high computational complexity and high power consumption cannot be handled by them. Neuromorphic processors, on the other hand, leverage spiking neural network (SNN) which has been shown to be less computationally complex and consume less power, making them suitable for such applications. Also, most robot arms unlike living agents that combine different sensory data to accurately perform a complex task, use uni-modal data which affects their accuracy. Conversely, multi-modal sensory data has been demonstrated to reach high accuracy and can be employed to achieve high accuracy in such robot arms. This paper presents the study of a multi-modal neurorobotic prosthetic arm control system based on recurrent spiking neural network. The robot arm control system uses multi-modal sensory data from visual (camera) and electromyography sensors, together with spike-based data processing on our previously proposed R-NASH neuromorphic processor to achieve robust accurate control of a robot arm with low power. The evaluation result using both uni-modal and multi-modal input data show that the multi-modal input achieves a more robust performance at 87%, compared to the uni-modal.

Abderazek Ben Abdallah , Zhishang Wang, K. N. Dang, Masayuki Hisada, '' Lacquering Robot System [漆塗りロボットシステム],' 特願2024-056380 (Macrh 29, 2023)
The 'Self-Controlled Urushi Robot Hand Painting System' is a cutting-edge robotic system designed to meet the evolving needs of modern Urushi painting. It skillfully manages a wide range of customer orders, interprets request formats such as images, text, or speech, and effectively handles the Urushi painting process. The system also ensures efficient task allocation based on robot hand 's availability and evaluates the adequacy of materials. It includes a blockchain platform that enables secure Urushi trading by generating digital certificates for the authenticity, preserving the integrity of traditional Urushi craftsmanship in a modern technological environment. In addition, robot hand intelligently monitors material inventory and autonomously procures the needed material by either using its profits or debiting the necessary amounts from linked bank accounts.