Myoelectric control systems—A survey

Using electromyogram (EMG) signals to control upper-limb prostheses is an important clinical option, offering a person with amputation autonomy of control by contracting residual muscles. The dexterity with which one may control a prosthesis has progressed very little, especially when control- ling multiple degrees of freedom. Using pattern recognition to discriminate multiple degrees of freedom has shown great promise in the research literature, but it has yet to transition to a clinically viable option. This article describes the pertinent issues and best practices in EMG pattern recognition, identifies the major challenges in deploying robust control, and advocates research directions that may have an effect in the near future.

Electromyogram pattern recognition for control of powered upper-limb prostheses: state of the art and challenges for clinical use.

Survey on human–robot collaboration in industrial settings: Safety, intuitive interfaces and applications

Recent advances in rehabilitation robotics suggest that it may be possible for hand-amputated subjects to recover at least a significant part of the lost hand functionality. The control of robotic prosthetic hands using non-invasive techniques is still a challenge in real life: myoelectric prostheses give limited control capabilities, the control is often unnatural and must be learned through long training times. Meanwhile, scientific literature results are promising but they are still far from fulfilling real-life needs. This work aims to close this gap by allowing worldwide research groups to develop and test movement recognition and force control algorithms on a benchmark scientific database. The database is targeted at studying the relationship between surface electromyography, hand kinematics and hand forces, with the final goal of developing non-invasive, naturally controlled, robotic hand prostheses. The validation section verifies that the data are similar to data acquired in real-life conditions, and that recognition of different hand tasks by applying state-of-the-art signal features and machine-learning algorithms is possible. Machine-accessible metadata file describing the reported data (ISA-Tab format)

/pdf/electromyography-data-for-non-invasive-naturally-controlled-hx7jp05mf2.pdf

Electromyography data for non-invasive naturally-controlled robotic hand prostheses

Natural control methods based on surface electromyography (sEMG) and pattern recognition are promising for hand prosthetics. However, the control robustness offered by scientific research is still not sufficient for many real life applications, and commercial prostheses are capable of offering natural control for only a few movements. In recent years deep learning revolutionized several fields of machine learning, including computer vision and speech recognition. Our objective is to test its methods for natural control of robotic hands via sEMG using a large number of intact subjects and amputees. We tested convolutional networks for the classification of an average of 50 hand movements in 67 intact subjects and 11 transradial amputees. The simple architecture of the neural network allowed to make several tests in order to evaluate the effect of pre-processing, layer architecture, data augmentation and optimization. The classification results are compared with a set of classical classification methods applied on the same datasets. The classification accuracy obtained with convolutional neural networks using the proposed architecture is higher than the average results obtained with the classical classification methods, but lower than the results obtained with the best reference methods in our tests. The results show that convolutional neural networks with a very simple architecture can produce accurate results comparable to the average classical classification methods. They show that several factors (including pre-processing, the architecture of the net and the optimization parameters) can be fundamental for the analysis of sEMG data. Larger networks can achieve higher accuracy on computer vision and object recognition tasks. This fact suggests that it may be interesting to evaluate if larger networks can increase sEMG classification accuracy too.

/pdf/deep-learning-with-convolutional-neural-networks-applied-to-invjgtz2v9.pdf

Deep Learning with Convolutional Neural Networks Applied to Electromyography Data: A Resource for the Classification of Movements for Prosthetic Hands

Assessment of operator stress induced by robot collaboration in assembly

The challenge of this work is to study the design and development of human-robot collaboration (HRC) in cellular manufacturing. Based on the concept of human collaborative design, four main design factors are being identified and developed in an active HRC prototype production cell for cable harness assembly. Human collaborative design aims to optimize the system design for the advantage of collaboration between human and robots based on human considerations. Task modeling approach is developed to study and analyze the task in order to identify the collaboration tasks to develop the collaboration planning. In the collaboration safety development, five safety designs, cover both hardware and control design, are proposed and developed in the prototype system. Risk assessment is conducted to verify the safety design. Two main experiments were conducted as preliminary study to investigate mental workload in HRC. A multimodal information support system is developed in the study of man-machine interface in this work to provide a comprehensive human-robot interface to facilitate human operator. The system performance evaluation had proven the improvement of prototype production cell with HRC design for cellular manufacturing.

/pdf/human-robot-collaboration-in-cellular-manufacturing-design-2cg9o1bpfw.pdf

Human-robot collaboration in cellular manufacturing: Design and development

This paper describes a new real-time learning method for the development of a robust motion discriminating method from an EMG signal, to adjust to the change in user's characteristics. This method is done under the assumptions that the input motions are continuous, and the teaching motions are ambiguous in nature, therefore, automatic addition, elimination and selection of learning data are possible. Applying our proposed method, we conducted experiments to discriminate eight forearm motions, with the results, a stable and highly effective discrimination rate was achieved and maintained even when changes occurred in user's characteristics.

Real-time Learning Method for Adaptable Motion-Discrimination using Surface EMG Signal

We have developed a hand rehabilitation system for patients suffering from paralysis or contracture. It consists of two components: a hand rehabilitation machine, which moves human finger joints with motors, and a data glove, which provides control of the movement of finger joints attached to the rehabilitation machine. The machine is based on the arm structure type of hand rehabilitation machine; a motor indirectly moves a finger joint via a closed four-link mechanism. We employ a wire-driven mechanism and develop a compact design that can control all three joints (i.e., PIP, DIP and MP ) of a finger and that offers a wider range of joint motion than conventional systems. Furthermore, we demonstrate the hand rehabilitation process, finger joints of the left hand attached to the machine are controlled by the finger joints of the right hand wearing the data glove.

Development of hand rehabilitation system for paralysis patient – universal design using wire-driven mechanism –

Our research aims to design and develop a safety strategy for a human–robot collaboration system. Although robotic assistance in a cellular manufacturing system is promising, safety is the uppermost consideration before it can be materialized. Five main safety designs are developed in this work. (i) Safe working areas for humans and robots. (ii) To control the behavior of the robot based on the collaboration requirements, light curtains defined safe collaborative working zones. (iii) Additionally, the robot system was developed using safe mechanical design and Dual Check Safety control strategies in terms of robot speed and travel area to minimize collaboration risks. (iv) A vision system using IP cameras was developed to monitor operator safety conditions by measuring the body posture and position of the operator. (v) The operation control system coordinated the collaborative flow between the operator and robot system. Apart from these developments, risk assessments were conducted to evaluate the safety ...

Ryu Kato

Papers

Assessment of operator stress induced by robot collaboration in assembly

Human-robot collaboration in cellular manufacturing: Design and development

Real-time Learning Method for Adaptable Motion-Discrimination using Surface EMG Signal

Development of hand rehabilitation system for paralysis patient – universal design using wire-driven mechanism –

Safety Strategy for Human-Robot Collaboration : Design and Development in Cellular Manufacturing