Path planning is the kernel problem of the robot technology area. In this paper, the grid method is used to make environmental modeling, Since the Genetic Algorithm (GA) has its immanent limitations and the Simulated Annealing (SA) Algorithm has the advantages in some aspects, combined these two algorithms together just achieve the perfection. In view of this, a hybrid of GA and SA (GA-SA Hybrid) is proposed in this paper to solve path planning problem for mobile robot. The algorithm making the crossover and mutation probability adjusted adaptively and nonlinearly with the completion time, can avoid such disadvantages as premature convergence. The new algorithm has better capability of searching globally and locally. The simulation results demonstrate that the proposed algorithm is valid and effective.

In this paper, a novel blind separation approach using wavelet and cross-wavelet is presented. This method extends the separate technology from time-frequency domain to time-scale domain. The simulation showed that this method is suitable for dealing with non-stationary signal.

Learner beliefs may promote or inhibit language learning. This experimental study explores metacognitive beliefs of EFL postgraduates majoring in science and technology (China), examining whether metacognitive beliefs change according to experimental conditions. The quantitative data comprise the responses of 90 postgraduates to a questionnaire on EFL learners' metacognitive beliefs (QELMB) at pre-test and post-test. The total metacognitive beliefs scores and the related twelve subscales were analyzed using SPSS version 15. The results indicate that some metacognitive beliefs tend to change with the experimental conditions while others show no significant change. The findings from the semi-structured interviews confirm the quantitative results, suggesting that metacognitive beliefs are vital to the development of language proficiency.

In this paper, a novel sparse component multi-resolution independent component analysis is presented. This method separates mixed images based on quadratic function of sparse component coefficient. The quadratic function can be interpreted as the time-frequency function or time-scale function. The performance of the algorithm is evaluated by using noisy mixed images data. Experimental results show that the method is feasible.

In this paper, a novel signal blind separation using adaptive multi-resolution independent component analysis based on sparse component is presented. This method separates mixed signal based on quadratic function and sparse representation. The quadratic function can be interpreted as the time-frequency function or time-scale function, or other. The sparse expression is the original signal through the dictionary to get their coefficients. Most of the coefficients is very small, close to zero, can greatly save separate computing time. At the same time this method can filter out the noise. The argorithm extends the separate technology from time-frequency domain to sparse mutil-resolution domain. The experimental result showed the method can be effective separation of mixed signals. And it shows that the method is feasible.