Gh. Lazea

Classification of surface electromyographic signals for control of upper limb virtual prosthesis using time-domain features

The development of a training system in the field of rehabilitation has always been a challenge for scientists. Surface electromyographical signals are widely used as input signals for upper limb prosthetic devices. The great mental effort of patients fitted with myoelectric prostheses during the training stage, can be reduced by using a simulator of such device. This paper presents an architecture of a system able to assist the patient and a classification technique of surface electromyographical signals, based on neural networks. Four movements of the upper limb have been classified and a rate of recognition of 96.67% was obtained when a reduced number of features were used as inputs for a feed-forward neural network with two hidden layers.

State estimation based on Kalman filtering techniques in navigation

This paper tackles the problem of the position measurement and estimation techniques in the robot navigation field based on Kalman filters. It presents the problem of the position estimation based on odometric, infrared and ultrasonic measurements. Further on deals with the theoretical and practical aspects of the state estimation based on Kalman filtering techniques. From the wide range of derivatives of the Kalman filtering technique there are detailed the extended Kalman filter and the one based on unscented transformation. In the second part of the paper is concluded with the results of the comparison between the different filtering algorithms and the further perspectives regarding this subject.

3D graphical simulation of an articulated serial manipulator based on kinematic models

This paper presents a portable 3D graphical simulator for a Fanuc M-6iB/2HS articulated robot. The mechanical structure is modeled using OpenGL functions implemented in Qt Framework. The robot motions are simulated based on the direct and inverse kinematics equations also presented in this paper. The simulator features include quick motion, step-by-step simulation, 3D scene control functions, objects selection functions and motion algorithms like joint interpolated motion and linear interpolated motion.

Hierarchical myoelectric control of a human upper limb prosthesis

Myolectric control is nowadays the most used approach for electrically-powered upper limb prostheses. The myoelectric controllers use electromyographic (EMG) signals as inputs. These signals can be collected from the skin surface using surface EMG sensors, or intramuscular, using needle sensors. No matter which method is used, they have to be processed before being used as controller inputs. In this paper, we present an algorithm based on an autoregressive (AR) model representation and a neural network, for EMG signal classification. The results have shown that combining a low-order AR model with a feed-forward neural network, a rate of classification of 98% can be achieved, while keeping the computational cost low. We also present a hierarchical control architecture and the implementation of the high-level controller using Finite State Machine. The solution proposed is capable of controlling three joints (i.e. six movements) of the upper limb prosthesis. The inputs of the high-level controller are obtained from the classifier, while its outputs are applied as input signals for the low-level controller. The main advantage of the proposed strategy is the reduced effort required to the patient for controlling the prosthetic device, since he only has to initiate the movement that is finalized by the low-level part of the controller.

An OpenGL Application for Industrial Robots Simulation

This paper presents a Visual C++ and OpenGL application for 3D simulation of the serial industrial robots. To develop this application we started from the forward kinematics of the robot taken into consideration. The functions implemented in the source code are able to calculate the position and orientation of each robot joint, including the position and orientation of the robot gripper. With the help of the OpenGL functions, the application is able to draw and simulate the 3D kinematic scheme of the robot. In addition, the application has a calculus module where the gripper position can be determined using particular values for the robot joints positions or orientations

An adaptive PID learning controller for periodic robot motion

We propose an adaptive PID learning controller which combines an adaptive PID feedback control scheme and a feedforward input learning design for the case of a periodic robot motion. The adaptive PID controller can overcome the transient response of the robot dynamics while the feedforward learning controller stands for computing the desired actuator torque needed for the nonlinear dynamics compensation in steady state. All the error signals that could appear in the learning control system are bounded, and the robot motion trajectory converges to the desired one asymptotically. On the other hand, the developed adaptive PID learning controller is compared with the fixed PID learning controller from the stability point of view to assure the gain bound, performance of tracking and, the most important, the convergence rate of the learning system.

Video based control of a 6 degrees-of-freedom serial manipulator

This paper presents a video based control system for a Fanuc M-6iB/2HS articulated robot. The system uses a CMUCam3 video camera connected to PC. The industrial robot is controlled via the TCP/IP protocol using a custom simulation software created in previous researches for the industrial robot. The simulation software implements additional classes designed to control and monitor the video camera. The paper presents in detail the calibration and testing stages. The system is capable to detect cylindrical objects of any stored color and is able to determine their position in the working space of the robot.

Industrial robot controller using miniature computers

This paper presents the conceptual design and the experimental results regarding the development of an industrial robot wireless controller using miniature computers. The industrial robot wireless controller is a part of a bigger project which has as main goal the development of a low-cost industrial robot simulation system. In this paper we describe the general hardware and software architecture of the controller, together with the preliminary experimental results.

Laser and vision based map building techniques for mobile robot navigation

We presents some preliminary results of the ongoing research with the final goal of building an autonomous mobile robot. To achieve this scope the mapping problem is an ineluctable one. This paper presents a visual mapping system which detects the same Speeded Up Robust Features (SURF) on the stereo pair images in order to obtain three dimensional point clouds at every robot location. The algorithm tracks the displacement of the identical features viewed from different positions to get back the robots positions. The Iterative Closest Point (ICP) algorithm is used to register the obtained landmarks in the feature based map of the entire environment. Also a mapping algorithm based on the laser system is presented which can detect the dynamic objects that are present in the robots field. The results of an indoor office environment experiments are shown.

Myoelectrical signal classification for the hierarchical control of a human hand prosthesis

Most electrically powered upper limb prostheses are myoelectrically controlled. The myoelectric controllers use surface electromyographical signals as inputs. These signals, collected from the surface of the skin, have to be preprocessed before being used as inputs for the controller. In this paper we present a classifier for surface electromyographical signals based on an autoregressive (AR) model representation and a neural network, and the higher level of the hierarchical controller implemented using Finite State Machine. The results had shown that using a low order autoregressive model combined with feed forward neural networks achieves a rate of classification of 91% while keeping the computational cost low. Using the hierarchical controller, the necessary effort to control the prosthesis by the patient is reduced since the patient only have to initiate the movement which is finalized by the low level part of the controller. The inputs of the high level controller are obtained from the classifier. The outputs of the high level controller are applied as inputs to the low level controller.