Research And Development Of Control Technology Of Visual Guided Grasping Robot

Target workpiece "Pick and place" movement is one of the most common action inindustrial robot automatic production line,and it is supporting technology of pallet sorting,handling, assembly, and so on.SCARA robot is very applicable to grasp due to its horizontalhigh flexibility, high vertical stiffness,and small volume,light quality,simple structure,highspeed and precision.Traditional teaching robot requires highly structured work environment,which has manybig limitations.Therefore,the research of visual guide grasp robot control technology toimprove the robots’ intelligence.The main research contents of this article are as follows:Visual guide grasp robot system modeling is studied.Combining with industrial intelligentcamera and a conveyor belt to build visual guide grasp robot system.System modeling mainlyincludes the SCARA robot kinematics modeling,industrial intelligent camera modeling,endtools modeling,industrial intelligent camera position relationship modeling between theconveyor belt and robots.Mathematical modeling was carried out on the system is the basis ofrealizing the follow-up trajectory tracking control and visual control.SCARA robot trajectory tracking control.This part improvement of sliding modecontroller of the SCARA robot is reached in view of the chattering of sliding model variablestructure and reaching the sliding surface in the process of reaching speed problems.Wedesign a fast nonsingular terminal sliding mode surface. Then, eliminating the chattering ofsliding mode control problem by using hyperbolic tangent function instead of the traditionalsymbol function.The problem of large torque due to the large initial state error is solvedthrough the adaptive fuzzy control by adjusting the parameters of the reaching law.Research of dynamic target grasping method based on the visual guide SCARA grasprobot system. a kind of dynamic target tracking method called "visual system+SCARA robot+conveyor belt" is designed.The grasping of dynamic target workpiece with the conveyorbelt moving is realized through industrial intelligent camera above the conveyor beltcontinuous access into the conveyor belt. The location of workpiece, combined with theconveyor belt speed encoder position feedback dynamic target workpiece as the currentposition, while planning SCARA robot trajectory in real-time.Image jacobian estimation algorithm based on the unscented kalman filtering.In order torealize the accurate placement for the grab workpiece,a visual servo control strategy is uesd inthis paper.For monocular vision, the Eye-in-Hand installation of industrial intelligent camerais utilized. An image jacobian estimation algorithm based on unscented kalman filtering isproposed in this paper.unscented kalman filter adopts the probability distribution of certainedsample points approximating nonlinear systems to approach nonlinear system. Compared withthe traditional kalman filter and extended kalman filter,the proposed algorithm improved theonline estimation precision of the image jacobian matrix effectively.