Advanced Materials Research Vols. 706-708

Abstract: The gait phase determination is the basis of the control of intelligent prosthesis. The prosthetic usually determines the gait phase according to its own motion parameters, which will cause the asymmetry gait of the prosthesis wearer during level walking due to the lack of contralateral motion information. By using of a variety of sensors installed in the healthy leg and prosthesis, the gait phase determination method based on echo control can achieve good symmetry, but at the same time also made calculation difficult. Based on the existing echo method, a gait phase determination method by utilize of the contralateral motion parameters is proposed. According to gait test data collected from healthy subjects, motion parameters and their relation can be obtained. Then the proposed method can determine its gait phase and contralateral phase. Preliminary experiment is conducted and verified the effective of the method.

Abstract: Wireless Sensor Networks consisting of nodes with limited power are deployed to collect and distribute useful information from the field to the other sensor nodes. Energy consumption is a key issue in the sensor’s communications since many use battery power, which is limited. In this paper, we describe a novel energy efficient routing approach which combines swarm intelligence, especially the ant colony based meta-heuristic, with a novel variation of reinforcement learning for sensor networks (ARNet). The main goal of our study was to maintain network lifetime at a maximum, while discovering the shortest paths from the source nodes to the sink node using an improved swarm intelligence. ARNet balances the energy consumption of nodes in the network and extends the network lifetime. Simulation results show that compared with the traditional EEABR algorithm can obviously improve adaptability and reduce the average energy consumption effectively.

Abstract: It is important for the control system analyst to understand the complete relationship of the complex-frequency representation of a linear system, the poles and zeros of its transfer function, and its time-domain response to step and other inputs. In such areas as signal processing and control, many of the analysis and designed collations are done in the complex-frequency plane, where a system model is represented in terms of the poles and zeros of its transfer functions.

Abstract: The multi-sensor data fusion widely applied to military and industry areas is a new technique developed in recent years. It can avoid the limitations of a single sensor and obtain more information, improving the recognition ability. This paper analyzes the uncertainties in the traditional earlier period diagnosis in induction motors based on the single parameter and introduces the idea of using multi-sensor data fusion to handle these uncertainties. In a fusion system several parameters will be fused according to the D-S evidential fusion algorithm in order to identify accurately the earlier period faults of the induction motors. Practical diagnostic examples show that the fault diagnostic accuracy and confidence are markedly promoted by using the multi-sensor data fusion.

Abstract: This paper designs the multilayer feed-forward neural network based on the immune genetic algorithm to solve the problem that BP algorithm is prone to get the local minimum in the failure diagnosis system. It is of both the learning ability and robustness of the neural network, as well as the strong global random searching ability of the immune genetic algorithm. The simulation results indicate the neural network can fulfill failure diagnosis of the complicated production better.

Abstract: Chip mounters are the important equipment on Surface Mounting Technology (SMT) production line, and with continuous improvement of the requirement on the mounting speed and mounting accuracy of micro-electronic components, high-speed and high-precision chip mounters have become important goals pursued by many companies and manufacturers of SMT equipment. As there is great difference in the shape, size, shape and number of pins of the components to be mounted by chip mounters, higher technical requirements on the mounting components identification and visual system must be put forward. The traditional mounter vision system controls and handles the light source system and the image acquisition system by personal computer (PC). However, in the practical application, the visual system has disadvantages such as large volume and high cost, which will lead to the shortcomings of bulky volume of the production equipment, inconvenient assembling, expensive price, difficult upgrading, etc. In this paper, to overcome these deficiencies of mounter vision system, based on the advantages of embedded system such as small size, fast response, reliable operation and relatively low prices, etc., the application of Advanced RISC Machines (ARM) embedded system in SMT machine vision system is researched, and the whole system and the module design is done with the modular design thought, which is conductive to the development and later maintenance, updating, and upgrade of the system.

Abstract: A new automatic data acquisition method, which uses a combination of single chip microcomputer and data acquisition card, was applied in this system. Single chip microcomputer captured the signal of photoelectric switches and delivered the information to upper computer by serial communication. The velocity of the moving load was obtained by photoelectric switch signal. The data acquisition program of upper computer was written by MATLAB instrument control toolbox and data acquisition toolbox. The data acquisition function of data acquisition card was triggered by the upper computer automatically. Testing data show that this system could automatic collect data of test bench stably. This system also shows high efficiency and reliability.

Abstract: Aiming at the problem of signal attenuation seriously, a block averaging pre-processing (BAP) differential coherent acquisition algorithm is proposed. BAP algorithm can be used to suppress interference signals and improve the stability of the receiver, but it is still faced with navigation signal transitions problem. Differential coherent algorithm can compensate for square loss caused by non-coherent accumulation. Monte Carlo simulation experiments show that the algorithm is more effective in the same carrier-to-noise ratio and same data length conditions.

Abstract: Based on single-chip microcomputer(SCM) control technology, designed a set of new brake automatic cooling control system. Through the hardware and software design of the system, and simulation test, proved that the system is reliable. The system can solve the problem of truck brake failure effectively.

Abstract: This paper uses biped walking robot as the research object, and designs robots original system, based on the requirements of Biped Walking Robot Competition of China. According to the biped walking robots characteristics of multi-joints, many degrees of freedom, multivariable, strong coupling and nonlinearity [, we can build system model using the Denavi - Hartenberg coordinate, describe the system model by the homogeneous coordinate transformation theory, and then plan on system gait based on ZMP stability . Finally, we can solve for the joint trajectory of the system by using computer-aided software.