Three-freedom-degree space mechanical arm motion planning method based on learning generalization mechanism

Abstract

The invention discloses a three-freedom-degree space mechanical arm motion planning method based on a learning generalization mechanism. Self motion characteristics obtained through training of a mechanical arm are adopted to build a motion sample database, through intelligent screening, an optimal learning sample is selected for learning, the mechanical arm generalizes motion characteristics of anew space target, and finally, according to the generalized space characteristics, joint motion of the mechanical arm is planned; the mechanical arm can carry out screening learning on the training motion of the mechanical arm so as to generalize motion of the new target, and capacity of cognition and learning generalization of the mechanical arm can be achieved; and the mechanical arm has the obstacle avoidance characteristic and has the good environment adaptation.

Description

technical field [0001] The invention belongs to the technical field of motion planning of a space manipulator, and in particular relates to a motion planning method of a manipulator capable of automatically avoiding obstacles. Background technique [0002] The multi-degree-of-freedom space manipulator has great application value in real life, but the complex structure of the manipulator and the large amount of motion planning calculations restrict the development and application of the manipulator. Traditional manipulator motion planning mainly refers to motion trajectory planning, which is mainly divided into two categories: joint space trajectory planning and Cartesian space trajectory planning. The former mainly uses polynomial interpolation methods in the traditional way, and the latter mainly uses space straight lines or space arcs. and other planning methods. Due to the multi-degree-of-freedom space characteristics of the manipulator itself, these methods not only mak...

AI Technical Summary

Problems solved by technology

Due to the multi-degree-of-freedom space characteristics of the manipulator itself, these methods not only make the planning and calculation complex in practical application, but also recalculate each pose of the manipulator during the movement process for each new target, especially when the manipulator When there are obstacles between the moving target point, the traditional method is more difficult to plan the movement of the robotic arm, so the traditional method has defects such as complex calculation, low intelligence, and poor adaptability

[0003] The current mainstream motion planning method is to use the open source motion planning library (OpenMotion Planning Library, hereinafter referred to as OMPL) developed by Rice University in the United States. Due to the high latitude characteristics of the space manipulator, the topology of its configuration space has fundamentally changes, most of the motion planning methods on the plane are no longer applicable

At present, the probabilistic road map method PRM and the rapid expansion random tree method RRT are mainly suitable for high-latitude motion planning in OMPL. These two algorithms are based on random sampling, although the planning algorithm based on sampling is very fast. , but there is a huge defect, that is, due to the randomness of the algorithm, the planning result cannot be predicted, and the result of each planning is different, so that the movement after planning is not stable

Images