Gesture tracking method based on population randomly-scrambled multi-target genetic algorithm

Abstract

The invention provides a gesture tracking method based on a population randomly-scrambled multi-target genetic algorithm. The method includes the following steps that arm skeleton information is acquired; a mechanical arm motion model is established; multi-target genetic algorithm energy and trajectory optimization is conducted; and gesture tracking is performed. According to the gesture tracking method based on the population randomly-scrambled multi-target genetic algorithm, the population randomly-scrambled multi-target genetic algorithm is introduced to achieve energy optimization, final position accuracy optimization and human body arm tracking trajectory optimization. Experimental results show that the system man-machine interaction is good, energy consumption of a mechanical arm is low, tail-end accuracy is high, and the human body arm tracking motion effect is good.

Description

technical field [0001] The invention relates to a gesture tracking method based on population random disorder multi-objective genetic algorithm, which belongs to the field of human-computer interaction automatic control of manipulators. Background technique [0002] With the development of technology, the importance of the interaction between humans and machines is increasing day by day. Vision-based gesture tracking technology has become an important part of human-computer interaction research, especially the use of this technology to effectively combine real human arms with simulated robotic arms to achieve harmonious, friendly, efficient and smooth human-computer communication. High-performance simulated robotic arms have been widely used in robotics, medical rescue, resource exploration, industry, teaching, and dangerous goods handling. Find ways to improve the movement of the robotic arm, increase its speed and accuracy, and ensure the safety of the robotic arm's trajec...

AI Technical Summary

Problems solved by technology

[0003] The existing operation mode of the mechanical arm is mainly operated by the handle or the rocker. The operation command cannot directly control the displacement and speed of the overall movement of the mechanical arm, making it difficult for the operator to complete complex movements without training, and the use flexibility is low. A certain delay, while the position accuracy cannot be guaranteed

Some studies have tried to use data gloves to operate the robotic arm. The data glove can accurately determine the position and posture of the operator's hand, but the existing practical robotic arm can achieve functions such as clamping without imitating hand movements, and it uses a large number of flexible devices. High cost, waste of functions

Images