Supervised non-contact mechanical arm demonstration method

Abstract

The invention discloses a supervised non-contact mechanical arm demonstration method, and belongs to robot demonstration technologies. According to the method, a mechanical arm contact operation demonstration task is executed when a mechanical arm tail end tool reaches a starting point position, the movement direction of the mechanical arm tail end tool is decided according to observation of a demonstrator, and meanwhile, a force signal value of a mechanical arm tail end six-dimensional force / torque sensor is observed. If the pose difference between the mechanical arm tail end tool and an operated object is different from an expected pose difference, the pose of the mechanical arm tail end tool is adjusted. In the adjusting process, if the contact force is too large and exceeds the bearing capacity of a mechanical arm or the object, the tail end pose of the mechanical arm needs to be adjusted to move in the direction ending contact. Otherwise, the pose of the mechanical arm tail end tool continues to be adjusted. When the demonstrator judges that the pose difference between the mechanical arm tail end tool and the operated object is the same as the expected pose difference, whether the demonstration task is finished or not is judged by continuously adjusting the pose of the mechanical arm tail end tool and data of the force sensor.

Description

technical field [0001] The invention belongs to the technical field of robot teaching, and in particular relates to a supervised and non-contact teaching method of a mechanical arm. Background technique [0002] Robots need to learn the mapping from input states to output actions when performing some tasks. The robot teaching technology is to record the state-action pairs of the expected robot behavior performed by the teacher, and use these data to generate a strategy to reproduce the teaching behavior. Robot teaching technology is usually divided into two stages: collecting samples and acquiring strategies based on samples. The most common way to collect samples is to drag the robot arm to collect the trajectory or map the motion of the human arm to the robot arm. However, such a method not only cannot achieve fine operation, but also in the teaching process of human dragging the robotic arm, the robotic arm is subjected to additional environmental forces imposed by huma...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that in the current teaching process of people dragging the mechanical arm, the mechanical arm is subjected to the extra influence exerted by the human being and cannot completely simulate the real application scene. The present invention provides a supervised non-contact The manipulator teaching method ensures that the teaching process is free from the influence of human force on the dynamics of the manipulator, and realizes the fine operation of the manipulator

Images