Absbact-Finding the body in uninterpreted sensory data is one of lhe fundamental competences to eonstrud the body repwsentation that influences on adaptabilities of the robat to the changes in the environment and the robot body. The invariance of sensations in self-observation seems a promising key information lo find the body. However, since each sensory attribute can he invariant only in the observation of a part of B e body, the robot should complementarily utilize the invariance of the multiple sensory attributes. In this paper, we propose B method of body-nonbody discrimination by complementarilg utilizing multiple sensory attributes based on a conjecture about the distribution of the variance of sensations for each ohsening posture, where it can be approximated by a midure of two Gaussian dislributions which are for obsenhg the body and the nonbody, respectively. By estimating the distribution, the robot can automatically find a discrimination hyperplane l o judge whether it observes its body in the " e n 1 observing posture. Simple experiments show the validity of the pmposed method.
A feasibility study has been conducted to determine whether mathematical models can be used for the numerical simulation of metal active gas (MAG) arc welding. In the present work, a three-dimensional, nonstationary thermal model for fillet welding is developed. The transient temperature distribution in the base metal is numerically analysed to estimate the molten pool size using a finite difference model based on the heat flow equation, and the theoretical configuration of the molten pool is calculated, taking account of the balance of gravity, surface tension, and arc pressure. The developed model can be applied to various welding processes such as multipass welding, lap welding, and welding with torch oscillation. To evaluate the validity of the model analysis, the calculated results are compared with experimental results for MAG welding. Good correspondence is demonstrated between experiment and calculation.
A feasibility study has been conducted to determine whether mathematical models can be used for the numerical simulation of MAG (Metal Active Gas) arc welding. In the present work, a three-dimensional (3-D), non-stationary thermal model for butt joint is developed. The transient temperature distribution in the base metal is numerically analyzed to estimate the molten pool size using a finite differential model based on the heat flow equation, and the theoretical configuration of the molten pool is calculated, taking account of the balance of gravity, surface tension and arc pressure. To evaluate the validity of the model analysis, the calculated results are compared with experimental results for MAG welding. Good correspondence is demonstrated between experiment and calculation. According to the software, the weld profile and temperature history in the work-piece for a set of process parameters is easily estimated. To predict the MAG arc welding phenomenon comprehensively, the developed welding process model has been linked to a deformation model. Accordingly, there is the possibility that the software is useful as an engineering tool in a manufacturing environment.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.