The dynamic modeling and simulation of a novel wall-climbing robot is presented. For the novel biped-wheel hybrid locomotion mechanism of the robot, its locomotion modes and typical state of motion are analyzed. Based on the description of the robots pose, the dynamic model for two typical states of motion, point turning in a flat surface and transition between two intersecting surfaces, is established. The equation for calculating adhesion force when the robot moves on arbitrary inclined surfaces is derived from the dynamic model. Simulations for the adhesion force are implemented with three typical examples. The results show that the tilt angle of the attachment surface and the motion direction of the robot have great influence on the adhesion force.
BP network is the most widely used of the neural net work model, but there are many problems of slow convergence speed and easily getting into the local minimum in the conventional BP algorithm. For this, an improved algorithm is proposed. Momentum term is added, steepness factors are introduced and adaptive learning rate adjustment factor is added. In the Matlab platform simulations are carried out by each improvement methods on the same BP neural network. The results show that: Convergence of improved BP network is decreased from 1000 to 49 and the error is decreased from 10-2 to 10-6. The convergence speed has been significantly improved and the error has been decreased. Using the synthesis improvement method effect is obvious and it provides a good theoretical basis for the practical application.
In order to overcome the low efficiency and precision disadvantages of steel plate position manual correction in the NC(numerical control) Flame Cutting, a high efficient steel plate position correction method is proposed by detecting the edge of steel plate images. Based on the development environment of OpenCV, the video of steel plate is captured based on the DirectShow Technology and the edge linear equation of steel plate is obtained by detecting the steel plate image edge with Canny algorithm and the angle between the equation and the x axis is obtained. The steel plate correction algorithm is designed, and based on the deviation angle of the steel plate, and rotation transformation of related processing images, the steel plate corrections is conducted automatically. The experimental testing of steel plate correction is done and the results show that the method can well done the steel plate correction and the error lower than 0.3 degree and can satisfy the flame cutting process requirement and have a great theory and economic value.
In order to develop an China-made high-performance thermal cutting numerical control(NC) system, the thermal cutting process characteristic is analyzed and an open, general-purpose thermal cutting 2D CNC system architecture is proposed, the hardware and software platforms are build, the functional modules of CNC system is developed and the machining test is done. The test results show that the CNC system is stable and powerful. it also has a high machining precision and can meet the need of thermal cutting.
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.