Song Aiguo

Virtual-Environment Modeling and Correction for Force-Reflecting Teleoperation With Time Delay

Virtual environment (VE) is an effective method to deal with time delay in teleoperation, but it depends heavily on the accuracy of the virtual model. In this paper, a novel approach is proposed to create and verify geometric and dynamic model of the remote physical environment. The geometric errors of the virtual model are corrected by overlaying the graphics over video images and also by fusing the position and force information from the remote. A sliding-average least-square algorithm is proposed to identify dynamic parameters of the remote environment, and the corresponding virtual-model parameters are updated online to keep equal to the real environment. The VE-based teleoperation system developed in our laboratory is described. Experimental results show that the relative errors of forces provided by the corrected VE are 2.82%, 2.22%, and 2.60%, respectively, with time delay of 5, 10, and 15 s. Results indicate that VE built by this method can provide the operator with appropriate predictive virtual forces with a certain time delayVirtual environment (VE) is an effective method to deal with time delay in teleoperation, but it depends heavily on the accuracy of the virtual model. In this paper, a novel approach is proposed to create and verify geometric and dynamic model of the remote physical environment. The geometric errors of the virtual model are corrected by overlaying the graphics over video images and also by fusing the position and force information from the remote. A sliding-average least-square algorithm is proposed to identify dynamic parameters of the remote environment, and the corresponding virtual-model parameters are updated online to keep equal to the real environment. The VE-based teleoperation system developed in our laboratory is described. Experimental results show that the relative errors of forces provided by the corrected VE are 2.82%, 2.22%, and 2.60%, respectively, with time delay of 5, 10, and 15 s. Results indicate that VE built by this method can provide the operator with appropriate predictive virtual forces with a certain time delay

Development of a force-assistant tele-rehabilitation system for the stroke

The goal of this study was to design a prototype of tele-rehabilitation system with force assistance that can be administered in both clinic and home therapy environments. A rehabilitation robotic device with its controller is designed to stretch and mobilize the impaired upper-limb safely and the interaction force and trajectory are sent to the server. The therapist in the hospital can control/monitoring the procedure and assess the effect of treatment remotely. Force-assistant tuning PID control scheme is introduced to adapt to different patients status. VR-based training interface is developed to inspire the subjects motivation. Result shows that the robotic arm can track the predictive trajectory on the whole although there are interaction forces between the upper-limb and the rehabilitation system. Compared with the traditional hospital-based therapy, this system is more cost-efficient and more convenient for both the stroke patients and the clinicians. Further work in this area would involve experimentation with real patients to determine the efficacy of this system in a therapy.

An improved genetic algorithm for sphericity error evaluation

Sphere is widely used in many engineering applications and its geometric error has a serious effect on the performance of mechanical parts. In this paper, a heuristic approach, that is an improved genetic algorithm (GA), is proposed to implement sphericity error evaluation. Then, the concepts of the minimum zone sphere (MZS), the minimum circumscribed sphere (MCS) and the maximum inscribed sphere (MIS) are clearly defined and their objective function calculation methods are formulated. Finally, the experiment results evaluated by different methods confirmed the proposed GA effectiveness. Compared with conventional evaluation methods, it has the advantages of simple algorithms, strong robustness and high precision. Also, it is a unified approach for other form errors evaluation.

Robust control with observer of internet-based teleoperation

Time delay reduces significantly ability of teleoperation. This problem becomes especially seriously in case of Internet-based teleoperation, when the time delay is variable. To overcome the trouble, time forward observer is used to predict the state of the slave. Moreover, force, position and velocity feedback is proposed to ensure that the system is robust asymptotically stable and transparent. The results of the experiment show effectiveness of this method.

Stochastic resonance in a single bistable neuron and its application

This paper investigated the response of a single bistable neuron to a weak periodic signal by the addition of the variable amount of noise. The results show that the neuron transduces the signal optimally when a certain amount of noise is present, i.e., stochastic resonance (SR). By the signal-to-noise ratio (SNR) measure, the neurons applicability and superiority for weak signal detection purposes are illustrated. The research enlarges the application scope of neuron to nonlinear signal processing.

Multi-mode haptic interaction technique and its application

The haptic interaction technology is an important human computer interaction technology widely used in virtual reality-based applications. Haptic feedbacks can be generated by haptic device and provided to human operators. The haptic feedback value is calculated by simulating the human haptic perception process of contacts with real objects in the virtual environment. The interactivity and telepresence of the virtual environment based human computer interaction can be greatly improved. Considering the aspects of virtual objects and haptic rendering, this paper summarizes the research situations of the technologies of haptic modeling of virtual objects and haptic rendering equipment, analyzes the characteristics of various modes of haptic interaction technology, and proposes the framework of a human-robot interaction-information perception system based on cloud computing. Besides, this paper briefly introduces the research achievements of the Instrument Science and Engineering lab of Southeast University. Finally, the future development trend of multi-mode haptic interaction technology is discussed.

Neural Network Control for Tele-rehabilitation Robot based on Variable Gain

The patients muscle spasm can cause the tele- rehabilitation robot system with force feedback to become instable and result in the slave unsmoothness of movement, which make the physical rehabilitative exercise inefficient during the tele-rehabilitative training. In order to guarantee the stability and reduce its fluctuation of the speed, a new method based on variable gain with back propagation neural network control was brought forward. With neural network adapting the control gains, not only the stability was guaranteed, but also the slave speed unsmoothness was lessened. The system can be used by sorts of rehabilitants and exhibits strong robustness. The simulation results demonstrate that this method is much more stable and smooth than that of the traditional control.