Search results

The purpose of this paper is to propose an accurate and practical imitation learning for robotics. The modified dynamic movement primitives (DMPs), global fitting DMPs (GLDMPs), is presented. Framework design, theoretical derivation and stability proof of GLDMPs are discussed in the paper.

Based on the DMPs, the hierarchical iterative parameter adaptive framework is developed as the hierarchical iteration stage of the GLDMPs to tune the designed parameters adaptively to extract richer features. Inspired by spatial transformations, the coupling analytical module which can be regarded as a reversible transformation is proposed to analyze the high-dimensional coupling information and transfer it to trajectory.

With the proposed framework and module, DMPs derive majority features of the demonstration and cope with three-dimensional rotations. Moreover, GLDMPs achieve favorable performance without specialized knowledge. The modified method has been demonstrated to be stable and convergent through inference.

GLDMPs have an advantage in accuracy, adaptability and practicality for it is capable of adaptively computing parameters to extract richer features and handling variations in coupling information. With demonstration and simple parameter settings, GLDMPs can exhibit excellent and stable performance, accomplish learning and generalize in other regions. The proposed framework and module in the paper are useful for imitation learning in robotics and could be intuitive for similar imitation learning methods.

The purpose of this paper is to investigate the influence of measurement uncertainties and the environment characteristics themselves on the desired field uniformity (FU) in reverberation chambers (RCs) by means of state-of-the-art global sensitivity analysis techniques. There are many quantities to describe the FU. The authors attempted to inspect many of the most important ones in two different orientations of the stirrer (horizontal, vertical).

Surrogate modelling techniques are involved to compute the Sobol’ indices efficiently with a modest number of required electromagnetic (EM) simulations. This can be only achieved if the behaviour of an appropriately chosen output quantity is predictable in such a way, which enables to extract useful information. Therefore, this choice should be made with extra care of the stochastic fluctuations, which have to be as low as possible. To this end, in this paper, various figures of merit are investigated.

This method can provide useful knowledge in the lower frequency range, where the ideal properties of the EM field in RCs cannot be established, and the importance of the setup parameters can vary from configuration to configuration.

Considering the current research tendencies related to RCs, the application of the method of Sobol’ indices to RCs is unique, which has only been done by the authors of this work so far. The main contribution presented in the paper is the thorough investigation of the effect of configuration parameters on the statistical properties of RCs through many output quantities describing the FU.