Applied Mechanics and Materials Vols. 556-562

Abstract: A digital PWM controller based on FPGA with the voltage feedback control mode is designed. The function blocks and the timing of the digital controller are introduced. The controller is applied in Buck circuits to verify the system function. The experimental results show that the designed controller has fast and strong real-time response speed with high reliability, which prove the correct theoretical analyses and feasibility of the proposed method.

Abstract: . When transporting the large and medium-sized stamping parts on automatic stamping production line, robotic end picking-up device (tooling) needs reasonable clamping positions to ensure smooth precision transmission and safe production. Based on UG platform, this paper has used the secondary development technique to develop a stamping process CAD system on automatic production line for robot manipulation with UG/Open API functions in Visual c++ 6.0 environment, the system can quickly analyze the stamping process and the clamping positions for robots and generate a three-dimensional entity mould, which effectively shorten the production line product development cycle, improve the quality and production efficiency of stamping parts.

Abstract: It is well known that the robot powered by pneumatic muscle actuator (PMA) which with highly nonlinear, time lag and time-varying characteristics is not easy to control. In this paper, we presented a hybrid fuzzy and PI controller for the upper-limb rehabilitation robot which was designed and built in our lab to assist patients post-stroke in the rehabilitation. The experimental results showed that controller can improve dynamic and static performance of the device, then it can control robot effectively to aid patients carrying on the systematized recovery training.

Abstract: For mobile robot localization in known environment, the 5^th-order Conjugate Unscented Particle Filter Monte Carlo Localization (CUPF-MCL) algorithm is proposed. CUPF-MCL combines the 5^th-order Conjugate Unscented Transform (5^th CUT) with Kalman Filter to generate more accuracy particle filter proposal distribution, calculating the transition density up to the 5^th-order nonlinearity. In simulation, the performance of CUPF-MCL is compared with that of dead reckoning, PF-MCL, EPF-MCL and UPF-MCL. Results show that CUPF-MCL improves the accuracy of localization.

Abstract: In this paper, a new sensorless control method is proposed for a permanent magnet linear synchronous motor based on Fuzzy sliding mode observer, which combines the advantages of sliding mode observer and Fuzzy controller respectively. The difference between the current estimated value and the actual current value is regarded as sliding mode function; sliding mode function (current error) and variation of the error are used as the input of fuzzy controller, and the width of the boundary layer as the output, adjusting the width of the boundary layer dynamically in real time. The simulation results show that Fuzzy sliding mode observer is able to find a balance between soft chattering and steady-state error, keep the system robustness and control precision.

Abstract: High frequency signal injection method is the most available method to achieve low speed and zero speed operations in mechanical sensor-less control of PMSM. The difficulty in HF signal injection method is how to pick up the HF components from the measured currents. Kalman filter is an adaptive filter and has been widely used in signal detections and processing. In this paper, a HF signal injection method using fuzzy Kalman filter for signal processing is proposed to realize the low speed operation in mechanical sensor-less control of PMSM. The Kalman filtering algorithm combined with a fuzzy inference system can efficiently adjust the covariance matrix of system measurement noise. Simulations and experiments are performed on PMSM, the results show that this method is a simple and direct signal processing method for HF signal injection method with very good stability and precision.

Abstract: The attitude estimation method of attitude heading reference system (AHRS) using an Extended Kalman Filter (EKF) with a filter tuning algorithm based on fuzzy controller is introduced.The AHRS uses inertial sensors and magnetometers to calculate its attitude. It is known that the attitude update using gyros are prone to diverge and hence the attitude error needs to compensate using accelerometers and magnetometers. In this paper, a Kalman filter model with a state variables represented by Modified Rodrigues Parameters (MRP) is presented to improve the computational efficiency and a model changing algorithm is used to make the filter more robust to acceleration and magnetic disturbances.If the AHRS measures any disturbances which are caused by movement of the vehicle, using fuzzy controller changes the filter gain .Simulation results show, EKF tuned by fuzzy controller is correct method that makes robust to disturbances more properly ,Rodrigues parameters can improve the computational efficiency..

Abstract: Existing Internet congestion control algorithms did not consider time-varying uncertainties. This paper presents a new congestion control algorithm based on adaptive technique and robust H_∞ approach, which can effectively compensate for the time-varying uncertainties. Simulation result shows that the proposed algorithm has good performance.

Abstract: This paper presents an adaptive robust controller for discrete-time systems with time-varying uncertainty and time-varying delay. The controller is designed based on the online parameter estimation and robust H∞ approach. Simulation result is given to verify the effectiveness of the proposed controller.

Abstract: For vehicle direct yaw moment control (DYC) ,the additional yaw moment decision method based on the fuzzy PI control and optimal allocation method of yaw moment based on quadratic programming are studied. Yaw moment control adopts hierarchical control method.The fuzzy PI controller and brake force optimization distributor are designed. The control method is verified through the Matlab/Simulink and CarSim co-simulation experiment.The results show that the control method can make the vehicle track the expected value better and improve the driving stability effectively.