Position Control of Cable-Driven Robotic Soft Arm Based on Deep Reinforcement Learning

Abstract

The cable-driven soft arm is mostly made of soft material; it is difficult to control because of the material characteristics, so the traditional robot arm modeling and control methods cannot be directly applied to the soft robot arm. In this paper, we combine the data-driven modeling method with the reinforcement learning control method to realize the position control task of robotic soft arm, the method of control strategy based on deep Q learning. In order to solve slow convergence and unstable effect in the process of simulation and migration when deep reinforcement learning is applied to the actual robot control task, a control strategy learning method is designed, which is based on the experimental data, to establish a simulation environment for control strategy training, and then applied to the real environment. Finally, it is proved by experiment that the method can effectively complete the control of the soft robot arm, which has better robustness than the traditional method.

Specifics of cruise tourism and features of creating a cruise tourism product

“…Taking into account the above, it should be emphasized that the primary features that determine the place of cruises in the specific structure of tourism are the physical and geographical characteristics of the environment in which the trip is carried out and the consumer properties of the corresponding tourist product. The technical features of the vehicles used for these purposes, which are both an object of accommodation and the provision of related services to passengers, are derived from primary -secondary features that are essential for the process of organizing this type of tourism [9,10]. Over the past decade, the basic consumer model of cruise tourism has multiplied across different market segments and has naturally integrated with other types of tourism.…”

Analysis of boundary conditions in design circuits influence on the shaft line operating state

“…3 Results The transverse vibrations natural frequency expression with one degree of freedom, with respect to the beam`s deflection from the load given, has the form [14]: (6) Lateral oscillations calculation allows you to check the ship`s shafting supports location, the length and selection of stern tube bearings elastic properties, their number, as well as mass and dimensions of the propeller. The ship`s shafting line lateral vibrations natural frequency value is affected by deflection parameter V0 at the propeller`s attachment point [15,16]. The greater deflection, the lower the shaft`s linear (6) transverse vibrations natural frequency.…”

Hierarchical control of soft manipulators towards unstructured interactions

“…And in their later work (Satheeshbabu et al, 2020), deep deterministic policy gradients (DDPGs) were used to implement control of the soft arm with continuous states and actions, and task space feedback was used to improve the ability to handle unknown load. Wu et al (2020) used DQN to implement a position controller of a soft arm in a vertical 2D plane. One of the problems to use reinforcement learning to control soft-bodied arms is that the data is hard to obtain.…”

“…In order to expedite training, Satheeshbabu et al (2019) used a mathematical model presented in Uppalapati and Krishnan (2021) to generate virtual training data. Wu et al (2020) used a similar method to generate training data. They used neural networks to train a model using real motion data of the soft arm, which can reduce model deviations caused by not considering non-linear factors in mathematical models.…”