Proximal strategy optimization-based high-speed gait generation method for underwater snakelike robot

Abstract

The invention discloses a proximal strategy optimization-based high-speed gait generation method for an underwater snakelike robot. The method comprises the steps of: building a training environment for reinforcement learning, wherein the training environment comprises a reset () function and a step () function; realizing a proximal strategy optimization algorithm in the training environment for reinforcement learning; and fitting the weight obtained by a strategic network by using a sine function as a primary function so as to convert the weight into a motion gait parameter of the underwatersnakelike robot. The gait generation method has the beneficial effects that in the gait generation method, a high speed is taken as an optimization target, and when the underwater snakelike robot optimizes obtained gait motion through the gait generation method, the forward motion speed of the underwater snakelike robot is higher than that when the obtained optimal gait motion is optimized througha traditional gait method; according to the gait generation method, in the optimization process, the gait can be found in a range greater than that of a snakelike gait equation, and the gait equationform obtained through the gait generation method is usually different from that of a snakelike gait equation, and is a novel motion gait.

Description

technical field [0001] The invention relates to a method for generating a high-speed gait of a robot. In particular, it relates to a method for high-speed gait generation of underwater snake-like robots based on proximal policy optimization. Background technique [0002] Due to the complex and changeable underwater environment, traditional underwater detection robots are usually difficult to deal with, and generally have the problem of short battery life. In recent years, robots based on bionics design have performed very well in adapting to complex environments. Their movement methods are derived from biological mechanisms, so the energy efficiency ratio is much higher than that of traditional robots. The design concept of the underwater snake-shaped robot researched by the present invention is derived from the biological snake, which has a redundant joint structure like the biological snake, has high flexibility, can move in various environments, and has high movement eff...

AI Technical Summary

Problems solved by technology

Aiming at the lack of motion control of the traditional gait of land snake robots in an unstructured environment, researchers at Carnegie Mellon University adopted an asynchronous advantage actor-critic (asynchronous advantage actor-critic, A3C) reinforcement learning algorithm under a distributed framework, The movement of the snake robot on the unstructured terrain is realized online and offline respectively, and the movement efficiency is improved by 40%, but the method still uses the prior knowledge of the traditional gait equation

Images