Research On Adaptive Control Of Attitude Self-perception Bidirectional Crawling Soft Robot

The soft robots have attracted much attention due to their strong adaptability,high degrees of freedom,and good safety.Therefore,since the "14th Five-Year Plan" in our country,the national government has vigorously supported the high-quality development of soft robots and related technologies,such as sensing,driving,and control.However,these soft robots still face challenges in perceiving complex environmental changes,accurate posture control,and limited motion patterns.In this thesis,we conducted research on integrated methods for self-perception and control of posture,taking into consideration materials,driving methods,structural design,sensing,and control methods.Firstly,to address the issue of difficulties in driving and extracting deformation characteristics of soft robots,this thesis proposes a flexible actuator based on liquid crystal elastomer with low-voltage driving and self-perception capabilities.Through comprehensive analysis of factors such as the stability,heating efficiency,and process stability of heating films,suitable flexible heating films are selected for experimental needs.To solve the problem of mutual interference between sensor signals and driving signals,this thesis integrates resistive flexible sensors on the back of the actuator using nano-imprint technology,enabling real-time detection of the bending posture of the flexible actuator.Meanwhile,the deformation performance,load-bearing capacity,driving mode,and response speed of the flexible actuator under different voltages are analyzed.Secondly,considering the issue of limited motion patterns of soft crawling robots,this thesis designs a soft robot that can crawl in both directions and has posture perception and motion mode switching capabilities.Through detailed analysis of the crawling motion patterns of caterpillars,this thesis sets graded friction forces on different parts of the robot to ensure seamless switching between bidirectional crawling and motion modes during the alternation of front and rear legs.In addition,through mechanical analysis of the structure of the soft robot,the motion mechanism of bidirectional crawling is elucidated,and a dynamic model of the flexible actuator of the robot is established based on the Euler-Lagrange kinematic equation,providing a theoretical basis for subsequent simulation analysis.Finally,considering that the driving of liquid crystal elastomer soft robots depends entirely on stress release,which leads to difficulties in posture control,this thesis proposes an integrated strategy based on self-perception and adaptive control of posture.The effectiveness of this strategy in posture control,anti-interference,and external pressure perception during the motion process of the flexible actuator and bidirectional crawling soft robot is demonstrated through simulation and experiments.Further tests of the robot’s application in complex environments,including coordinated motion in different planes and narrow pipelines,visual exploration with mounted cameras,and adjustment of front and rear leg speeds when encountering obstacles,are conducted.