scholarly journals A novel energy-motion model for continuous sEMG decoding: from muscle energy to motor pattern

2020 ◽  
Author(s):  
Gang Liu ◽  
Lu Wang ◽  
Jing Wang
Keyword(s):  
Real Time ◽  
Degrees Of Freedom ◽  
Motor Pattern ◽  
Human Hand ◽  
Motion Model ◽  
Real Time Control ◽  
Prosthetic Hands ◽  
Time Control ◽  
Energy Mode ◽  
Muscle Energy

Myoelectric prosthetic hands create the possibility for amputees to control their prosthetics like native hands. However, user acceptance of the extant myoelectric prostheses is low. Unnatural control, lack of sufficient feedback, and insufficient functionality are cited as primary reasons. Recently, although many multiple degrees-of-freedom (DOF) prosthetic hands and tactile-sensitive electronic skins have been developed, no non-invasive myoelectric interfaces can decode both forces and motions for five-fingers independently and simultaneously. This paper proposes a myoelectric interface based on energy allocation and fictitious forces hypothesis by mimicking the natural neuromuscular system. The energy-based interface uses a kind of continuous “energy mode” in the level of the entire hand. According to tasks itself, each energy mode can adaptively and simultaneously implement multiple hand motions and exerting continuous forces for a single finger. Also, a few learned energy modes could extend to the unlearned energy mode, highlighting the extensibility of this interface. We evaluate the proposed system through off-line analysis and operational experiments performed on the expression of the unlearned hand motions, the amount of finger energy, and real-time control. With active exploration, the participant was proficient at exerting just enough energy to five fingers on “fragile” or “heavy” objects independently, proportionally, and simultaneously in real-time. The main contribution of this paper is proposing the bionic energy-motion model of hand: decoding a few muscle-energy modes of the human hand (only ten modes in this paper) map massive tasks of bionic hand.

Real Time, Near-Optimal Trajectory Planning of an Omni-Directional Vehicle

2001 ◽  
Author(s):  
Tamás Kalmár-Nagy ◽  
Pritam Ganguly ◽  
Raffaello D’Andrea
Keyword(s):  
Real Time ◽  
Optimal Trajectory ◽  
Degrees Of Freedom ◽  
Dynamic Environments ◽  
Real Time Control ◽  
Uncertain Environments ◽  
Time Control ◽  
Innovative Method ◽  
Optimal Trajectory Planning ◽  
Three Degrees Of Freedom

Abstract In this paper, we discuss an innovative method of generating near-optimal trajectories for a robot with omni-directional drive capabilities, taking into account the dynamics of the actuators and the system. The relaxation of optimality results in immense computational savings, critical in dynamic environments. In particular, a decoupling strategy for each of the three degrees of freedom of the vehicle is presented, along with a method for coordinating the degrees of freedom. A nearly optimal trajectory for the vehicle can typically be calculated in less than 1000 floating point operations, which makes it attractive for real-time control in dynamic and uncertain environments.

Investigation of Different Approaches to Real-Time Control of Prosthetic Hands with Electromyography Signals

2021 ◽  
pp. 1-1
Author(s):  
Joao Olegario de O. De Souza ◽  
Marcos D. Bloedow ◽  
Felipe Rubo ◽  
Rodrigo M. De Figueiredo ◽  
Gustavo Pessin ◽  
...  
Keyword(s):  
Real Time ◽  
Real Time Control ◽  
Prosthetic Hands ◽  
Time Control ◽  
Electromyography Signals

scholarly journals Myoelectric digit action decoding with multi-label, multi-class classification: an offline analysis

2020 ◽  
Author(s):  
Agamemnon Krasoulis ◽  
Kianoush Nazarpour
Keyword(s):  
Real Time ◽  
Degrees Of Freedom ◽  
Real Time Control ◽  
Artificial Joints ◽  
Time Control ◽  
Offline Analysis ◽  
Prosthesis Control ◽  
Multi Class Classification ◽  
Prosthetic Finger ◽  
Finger Control

ABSTRACTThe ultimate goal of machine learning-based myoelectric control is simultaneous and independent control of multiple degrees of freedom (DOFs), including wrist and digit artificial joints. For prosthetic finger control, regression-based methods are typically used to reconstruct position/velocity trajectories from surface electromyogram (EMG) signals. Although such methods have produced highly-accurate results in offline analyses, their success in real-time prosthesis control settings has been rather limited. In this work, we propose action decoding, a paradigm-shifting approach for independent, multi-digit movement intent decoding based on multi-label, multi-class classification. At each moment in time, our algorithm classifies movement action for each available DOF into one of three categories: open, close, or stall (i.e., no movement). Despite using a classifier as the decoder, arbitrary hand postures are possible with our approach. We analyse a public dataset previously recorded and published by us, comprising measurements from 10 able-bodied and two transradial amputee participants. We demonstrate the feasibility of using our proposed action decoding paradigm to predict movement action for all five digits as well as rotation of the thumb. We perform a systematic offline analysis by investigating the effect of various algorithmic parameters on decoding performance, such as feature selection and choice of classification algorithm and multi-output strategy. The outcomes of the offline analysis presented in this study will be used to inform the real-time implementation of our algorithm. In the future, we will further evaluate its efficacy with real-time control experiments involving upper-limb amputees.

scholarly journals Myoelectric digit action decoding with multi-output, multi-class classification: an offline analysis

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Agamemnon Krasoulis ◽  
Kianoush Nazarpour
Keyword(s):  
Real Time ◽  
Degrees Of Freedom ◽  
Real Time Control ◽  
Artificial Joints ◽  
Time Control ◽  
Offline Analysis ◽  
Intent Prediction ◽  
Multi Class Classification ◽  
Prosthetic Finger ◽  
Finger Control

Abstract The ultimate goal of machine learning-based myoelectric control is simultaneous and independent control of multiple degrees of freedom (DOFs), including wrist and digit artificial joints. For prosthetic finger control, regression-based methods are typically used to reconstruct position/velocity trajectories from surface electromyogram (EMG) signals. Unfortunately, such methods have thus far met with limited success. In this work, we propose action decoding, a paradigm-shifting approach for independent, multi-digit movement intent prediction based on multi-output, multi-class classification. At each moment in time, our algorithm decodes movement intent for each available DOF into one of three classes: open, close, or stall (i.e., no movement). Despite using a classifier as the decoder, arbitrary hand postures are possible with our approach. We analyse a public dataset previously recorded and published by us, comprising measurements from 10 able-bodied and two transradial amputee participants. We demonstrate the feasibility of using our proposed action decoding paradigm to predict movement action for all five digits as well as rotation of the thumb. We perform a systematic offline analysis by investigating the effect of various algorithmic parameters on decoding performance, such as feature selection and choice of classification algorithm and multi-output strategy. The outcomes of the offline analysis presented in this study will be used to inform the real-time implementation of our algorithm. In the future, we will further evaluate its efficacy with real-time control experiments involving upper-limb amputees.

scholarly journals A geometric approach for follow-the-leader motion of serpentine manipulator

2019 ◽  
Vol 16 (5) ◽  
pp. 172988141987463 ◽  
Author(s):  
Haibo Xie ◽  
Cheng Wang ◽  
Shusen Li ◽  
Liang Hu ◽  
Huayong Yang
Keyword(s):  
Path Planning ◽  
Real Time ◽  
Obstacle Avoidance ◽  
Degrees Of Freedom ◽  
Mechanical Design ◽  
Computation Time ◽  
Geometric Approach ◽  
Real Time Control ◽  
Time Control ◽  
The Ideal

This article presents a geometric approach for path planning of serpentine manipulator for real-time control in confined spaces. Firstly, the mechanical design of a serpentine manipulator is introduced, and its kinematics is analyzed. As the serpentine manipulator usually has more than 10 degrees of freedom, the motion control and obstacle avoidance are difficult considering its inverse kinematics. Follow-the-leader is an ideal path planning method for serpentine manipulator, as the manipulator moves forward, all the sections follow the path that the tip of manipulator has passed, which simplifies the obstacle avoidance. The realization of follow-the-leader method is to find the new configurations of the manipulator that can fit the ideal path with small errors. In this article, a novel geometric approach for follow-the-leader motion is proposed to solve new configurations with high precision of location and less computation time. The method is validated through simulation and the deviation from the ideal path is analyzed, simulation results show that calculation time for per step is less than 0.5 ms for a serpentine manipulator with 10 sections. To verify the follow-the-leader method, a 13-degree-of-freedom serpentine manipulator system with 6 sections was built, and 12 magnetic rotary encoders were embedded into the universal joints to collect data of rotation angles of each section. Experimental results show that the manipulator can carry out follow-the-leader motion as expected in real time.

scholarly journals A novel energy-motion model for continuous sEMG decoding: from muscle energy to motor pattern

2020 ◽  
Author(s):  
Gang Liu ◽  
Lu Wang ◽  
Jing Wang
Keyword(s):  
Potential Energy ◽  
Motor Pattern ◽  
Training Data ◽  
Hand Movements ◽  
Coupling Mechanism ◽  
Real Time Control ◽  
Time Control ◽  
Single Finger ◽  
Energy Mode ◽  
Semg Signals

<p></p><p><i>Background:</i> <a>At present, the gesture recognition using sEMG signals requires vast amounts of training data or limits to a few hand movements. This paper presents a novel dynamic energy model that can decode continuous hand actions with</a> force information, by training small amounts of sEMG data.</p> <p><i>Method:</i> As activating the forearm muscles, the corresponding fingers are moving or tend to move (namely exerting force). The moving fingers store kinetic energy, and the fingers with moving trends store potential energy. The kinetic and potential energy of fingers is dynamically allocated due to the adaptive-coupling mechanism of five-fingers in actual motion. At this certain moment, the sum of the two energies is constant. We regarded energy mode with the same direction of acceleration of each finger, but likely different movements, as the same one, and divided hand movements into ten energy modes. Independent component analysis and machine learning methods were used to model associations between sEMG signals and energy mode, to determine the hand action, including speed and force adaptively. This theory imitates the self-adapting mechanism in the actual task; thus, ten healthy subjects were recruited, and three experiments mimicking activities of daily living were designed to evaluate the interface: (1) decoding untrained configurations, (2) decoding the amount of single-finger energy, and (3) real-time control.</p> <p><i>Results:</i>(1) Participants completed the untrained hand movements (100 /100, p < 0.0001). (2) The test of pricking balloon with a needle tip was designed with significantly better than chance (779 /1000, p < 0.0001).(3) The test of punching a hole in the plasticine on the balloon was with over 95% success rate (97.67±5.04 %, p <0.01).</p> <p><i>Conclusion: </i>The model can achieve continuous hand actions with force information, by training small amounts of sEMG data, which reduces trained complexity.</p><p></p>

scholarly journals A novel energy-motion model for continuous sEMG decoding: from muscle energy to motor pattern

2020 ◽  
Author(s):  
Gang Liu ◽  
Lu Wang ◽  
Jing Wang
Keyword(s):  
Potential Energy ◽  
Motor Pattern ◽  
Training Data ◽  
Hand Movements ◽  
Coupling Mechanism ◽  
Real Time Control ◽  
Time Control ◽  
Single Finger ◽  
Energy Mode ◽  
Semg Signals

<p></p><p><i>Background:</i> <a>At present, the gesture recognition using sEMG signals requires vast amounts of training data or limits to a few hand movements. This paper presents a novel dynamic energy model that can decode continuous hand actions with</a> force information, by training small amounts of sEMG data.</p> <p><i>Method:</i> As activating the forearm muscles, the corresponding fingers are moving or tend to move (namely exerting force). The moving fingers store kinetic energy, and the fingers with moving trends store potential energy. The kinetic and potential energy of fingers is dynamically allocated due to the adaptive-coupling mechanism of five-fingers in actual motion. At this certain moment, the sum of the two energies is constant. We regarded energy mode with the same direction of acceleration of each finger, but likely different movements, as the same one, and divided hand movements into ten energy modes. Independent component analysis and machine learning methods were used to model associations between sEMG signals and energy mode, to determine the hand action, including speed and force adaptively. This theory imitates the self-adapting mechanism in the actual task; thus, ten healthy subjects were recruited, and three experiments mimicking activities of daily living were designed to evaluate the interface: (1) decoding untrained configurations, (2) decoding the amount of single-finger energy, and (3) real-time control.</p> <p><i>Results:</i>(1) Participants completed the untrained hand movements (100 /100, p < 0.0001). (2) The test of pricking balloon with a needle tip was designed with significantly better than chance (779 /1000, p < 0.0001).(3) The test of punching a hole in the plasticine on the balloon was with over 95% success rate (97.67±5.04 %, p <0.01).</p> <p><i>Conclusion: </i>The model can achieve continuous hand actions with force information, by training small amounts of sEMG data, which reduces trained complexity.</p><p></p>

A New Real-Time Iteration Method Based on Geometry for Inverse Kinematics of Truck Mounted Concrete Pump

2013 ◽  
Vol 373-375 ◽  
pp. 2109-2113
Author(s):  
Long An Chen ◽  
Ying Jie Shen ◽  
Zhi Nan Mi
Keyword(s):  
Real Time ◽  
Iteration Method ◽  
Inverse Kinematics ◽  
Degrees Of Freedom ◽  
Variable Step Size ◽  
Real Time Control ◽  
Step Size ◽  
Time Control ◽  
Variable Step ◽  
Concrete Pump

A new iteration method based on geometry to solve the inverse kinematics for the boom system of truck mounted concrete pump, which is difficult to real-time control since its degrees of freedom are multiple-redundant, is presented. This method uses a variable-step size technique to approach the solution of the inverse kinematics, and uses geometry to determine how much angles of joint to change and its direction. Comparing with the traditional methods, this method is more suitable for real-time control of truck mounted concrete pump boom system without calculating the inverse matrix of jacobian. By the method the movement of boom will be safer and more stable when pumping concrete. Simulation results show that the new method has a fast convergence speed and good stability.

Design and real-time implementation of a decentralized sliding mode controller for twin rotor multi-input multi-output system

2017 ◽  
Vol 231 (1) ◽  
pp. 3-13 ◽  
Author(s):  
Farah Faris ◽  
Abdelkrim Moussaoui ◽  
Boukhetala Djamel ◽  
Tadjine Mohammed
Keyword(s):  
Real Time ◽  
Degrees Of Freedom ◽  
Sliding Mode ◽  
System Control ◽  
Sliding Mode Controller ◽  
Real Time Control ◽  
Strongly Coupled ◽  
Time Control ◽  
Output System ◽  
Twin Rotor

The article deals with a real-time implementation of a decentralized sliding mode controller applied to a twin rotor multi-input multi-output system, a system with 2 degrees of freedom, strongly coupled and its dynamic resembles that of a helicopter. The work is motivated by the fact that in the literature several control techniques have been proposed for the twin rotor multi-input multi-output system control without being applied to the system, and the considered authors presented just the simulation results. To control the vertical and horizontal positions of the twin rotor multi-input multi-output system, the system is decoupled into two subsystems, vertical and horizontal, controlled by two independent sliding mode regulators calculated from the mathematical models of vertical and horizontal subsystems, respectively. From the results of real-time control of the twin rotor multi-input multi-output system in stabilization and tracking modes, and performing robustness and disturbance rejection tests, the effectiveness of the suggested control scheme was proven.

Based on a Microcontroller the Real-Time Control System Design of Manipulator of Multiple Degrees of Freedom

2014 ◽  
Vol 722 ◽  
pp. 213-216
Author(s):  
Rui Yong Duan ◽  
Ji Ling Yan
Keyword(s):  
Control System ◽  
Real Time ◽  
Degrees Of Freedom ◽  
Time Synchronization ◽  
Synchronization Control ◽  
Control System Design ◽  
Real Time Control ◽  
Multiple Degrees Of Freedom ◽  
Time Control ◽  
Wireless Transmitter

Manipulator of multiple degrees of freedom has been widely used in all kinds of rescue and exploring robots. This paper regard a kind of 6 DOF manipulator controller as the design object.The data acquisition for WDD35D4 sensors is carried by the C8051F020 microcontroller and sent through a wireless transmitter NRF905 after encrypted.Then the wireless transmitter NRF905 of the manipulator receives decryption. Through data processing and the steering gear adjustment, the angle of linear potentiometer corresponds to the angle of steering by 1-1. Then it drives the manipulator to work out and simulate the same motion of mechanical arm controller, and make real-time synchronization control.

Sign in / Sign up

Export Citation Format

Share Document