Search

The joint hydraulic drive unit (HDU) serves as a pivotal element in enabling the high-performance movements of legged robots. Functioning as the conduit linking the oil source and the actuator, the hydraulic flow channel significantly impacts actuator performance. Hence, optimizing the HDU flow channel becomes imperative, enhancing not only HDU efficiency but also the overall system performance. This paper introduces a novel approach by aligning the hydraulic flow channel of the joint HDU with the arteriovenous layout of the cardiac vascular system, departing from the conventional machining flow channel model. Through simulations determining the optimal range of the vascular branch radius and angle, this study guides the design optimization of the joint HDU flow channel. With the primary optimization goal of reducing pressure loss, the study compares simulation outcomes of various flow channel models—linear, variable excessive radius, and the multidimensional Bessel curve—tailored to suit the arrangement specifics of the joint HDU. Further validating these designs, the flow channels are fabricated using additive manufacturing for experimental verification. The integration of simulation analyses and pressure loss testing reveals a remarkable reduction of over 40% in pressure loss for the bionic flow channel compared to the conventional machining form. This empirical evidence strongly substantiates the bionic flow channel's superior efficacy in pressure loss reduction. The findings presented herein offer valuable insights for the development of low-loss flow channels in joint HDUs, thereby presenting a new avenue for designing energy-efficient, high power-to-weight ratio legged robots. [ABSTRACT FROM AUTHOR]

SUMMARY: In this paper, mathematical models of kinematics, statics and inverse dynamics are derived firstly according to the mechanical structure of leg hydraulic drive system (LHDS). Then, all the above models are integrated with MATLAB/Simulink to build the LHDS simulation model, the model not only considers influence of leg dynamic characteristics on hydraulic system but also takes into account nonlinearity, variable load characteristics and other common problems brought by hydraulic system, and solves compatibility and operation time which brought by using multiple software simultaneously. The experimental results show the simulation model built in this paper can accurately express characteristics of the system. [ABSTRACT FROM AUTHOR]

Abstract Hydraulic drive mode enables legged robots to have excellent characteristics, such as greater power-to-weight ratios, higher load capacities, and faster response speeds than other robots. Nowadays, highly integrated valve-controlled cylinder, called hydraulic drive unit (HDU), is employed to drive the joints of these robots. However, various robot control issues exist. For example, during the walking process of legged robots, different obstacles are encountered, making it difficult to control such robots because the load characteristics of the ends of their feet change with the environment. Furthermore, although the adoption of HDU has resulted in high-performance robot control, the hydraulic systems of these robots still have problems, such as strong nonlinearity, and time-varying parameters. Consequently, robot control is very difficult and complex. This paper proposes an improved second-order dynamic compliance control system, impedance control, for HDU. The control system is designed to rectify the issues affecting the impedance control accuracy of the dynamic compliance serial–parallel composition between the HDU force control inner loop and the impedance control outer loop. Specifically, it consists of a compliance-enhanced controller and a feedforward compensation controller for the force control inner loop. Furthermore, the dynamic compliance composition of the inner and outer HDU control loops is rearranged. The results of experiments conducted indicate that the proposed method significantly improves the control accuracy compared to that of traditional force-based impedance control. Highlights • Key reasons affecting the accuracy of the traditional force-based impedance control. • Dynamic compliance composition of both force inner loop and impedance outer loop. • Control accuracy of the traditional force-based impedance control is improved. [ABSTRACT FROM AUTHOR]

In this paper, the mathematical expressions of a second-order matrix sensitivity analysis (SOMSA) is derived. This method has higher accuracy and requires less calculation works than previous analysis methods. Based on the SOMSA, when the traditional force-based impedance control is applied in leg hydraulic drive system (LHDS) of legged robots, the effects of parameter variations on control performance are studied by sensitivity dynamic analysis under nine working conditions. Then, combined with two measurable sensitivity indexes, the results of the sensitivity dynamic analysis are studied quantitatively. Finally, the above results are verified experimentally by using LHDS test platform. The conclusions contribute to the optimization of the LHDS structure and provide theoretical references for compensation control strategies of the traditional force-based impedance control. • New sensitivity analysis method called second-order matrix sensitivity analysis(SOMSA) is proposed. • Effects of parameter variations on control performance of force-based impedance control is studied. • Sensitivity dynamic analysis is conducted for twelve main parameters under nine working conditions. • Sensitivity dynamic analysis are studied quantitatively by combining with two measurable sensitivity indexes. [ABSTRACT FROM AUTHOR]