Deep reinforcement learning-based attitude motion control for humanoid robots with stability constraints

2020 ◽  
Vol 47 (3) ◽  
pp. 335-347 ◽  
Author(s):  
Qun Shi ◽  
Wangda Ying ◽  
Lei Lv ◽  
Jiajun Xie
Keyword(s):  
Attitude Control ◽  
A Priori ◽  
Control Process ◽  
Humanoid Robots ◽  
Content Type ◽  
Training Samples ◽  
Attitude Controller ◽  
Low Efficiency ◽  
The Stability ◽  
And Control

Purpose This paper aims to present an intelligent motion attitude control algorithm, which is used to solve the poor precision problems of motion-manipulation control and the problems of motion balance of humanoid robots. Aiming at the problems of a few physical training samples and low efficiency, this paper proposes an offline pre-training of the attitude controller using the identification model as a priori knowledge of online training in the real physical environment. Design/methodology/approach The deep reinforcement learning (DRL) of continuous motion and continuous state space is applied to motion attitude control of humanoid robots and the robot motion intelligent attitude controller is constructed. Combined with the stability analysis of the training process and control process, the stability constraints of the training process and control process are established and the correctness of the constraints is demonstrated in the experiment. Findings Comparing with the proportion integration differentiation (PID) controller, PID + MPC controller and MPC + DOB controller in the humanoid robots environment transition walking experiment, the standard deviation of the tracking error of robots’ upper body pitch attitude trajectory under the control of the intelligent attitude controller is reduced by 60.37 per cent, 44.17 per cent and 26.58 per cent. Originality/value Using an intelligent motion attitude control algorithm to deal with the strong coupling nonlinear problem in biped robots walking can simplify the control process. The offline pre-training of the attitude controller using the identification model as a priori knowledge of online training in the real physical environment makes up the problems of a few physical training samples and low efficiency. The result of using the theory described in this paper shows the performance of the motion-manipulation control precision and motion balance of humanoid robots and provides some inspiration for the application of using DRL in biped robots walking attitude control.

scholarly journals Turning Gait Planning Method for Humanoid Robots

2018 ◽  
Vol 8 (8) ◽  
pp. 1257 ◽  
Author(s):  
Tianqi Yang ◽  
Weimin Zhang ◽  
Xuechao Chen ◽  
Zhangguo Yu ◽  
Libo Meng ◽  
...  
Keyword(s):  
Inverted Pendulum ◽  
Center Of Mass ◽  
Translational Motion ◽  
Humanoid Robots ◽  
Inverted Pendulum Model ◽  
Straight Line ◽  
Pendulum Model ◽  
The Stability ◽  
And Control ◽  
Present Simulation

The most important feature of this paper is to transform the complex motion of robot turning into a simple translational motion, thus simplifying the dynamic model. Compared with the method that generates a center of mass (COM) trajectory directly by the inverted pendulum model, this method is more precise. The non-inertial reference is introduced in the turning walk. This method can translate the turning walk into a straight-line walk when the inertial forces act on the robot. The dynamics of the robot model, called linear inverted pendulum (LIP), are changed and improved dynamics are derived to make them apply to the turning walk model. Then, we expend the new LIP model and control the zero moment point (ZMP) to guarantee the stability of the unstable parts of this model in order to generate a stable COM trajectory. We present simulation results for the improved LIP dynamics and verify the stability of the robot turning.

scholarly journals A novel steady-state maintenance simulation framework for multi- information disseminations in crowd network

2020 ◽  
Vol 4 (3) ◽  
pp. 273-282
Author(s):  
Zhong Wang ◽  
Hongbo Sun ◽  
Baode Fan
Keyword(s):  
Steady State ◽  
Design Methodology ◽  
Information Dissemination ◽  
Simulation Platform ◽  
Simulation Framework ◽  
Source Information ◽  
Content Type ◽  
General Rules ◽  
The Stability ◽  
And Control

Purpose The era of crowd network is coming and the research of its steady-state is of great importance. This paper aims to establish a crowd network simulation platform and maintaining the relative stability of multi-source dissemination systems. Design/methodology/approach With this simulation platform, this paper studies the characteristics of “emergence,” monitors the state of the system and according to the fixed point judges the system of steady-state conditions, then uses three control conditions and control methods to control the system status to acquire general rules for maintain the stability of multi-source information dissemination systems. Findings This paper establishes a novel steady-state maintenance simulation framework. It will be useful for achieving controllability to the evolution of information dissemination and simulating the effectiveness of control conditions for multi-source information dissemination systems. Originality/value This paper will help researchers to solve problems of public opinion control in multi-source information dissemination in crowd network.

scholarly journals A Small Membrane Stabilizing Protein Critical to the Pathogenicity of Staphylococcus aureus

2020 ◽  
Vol 88 (9) ◽  
Author(s):  
Seána Duggan ◽  
Maisem Laabei ◽  
Alaa Abdulaziz Alnahari ◽  
Eóin C. O’Brien ◽  
Keenan A. Lacey ◽  
...  
Keyword(s):  
Staphylococcus Aureus ◽  
Iron Homeostasis ◽  
Control Strategies ◽  
Antibiotic Resistant ◽  
Content Type ◽  
Promising Target ◽  
Significant Impairment ◽  
The Stability ◽  
Conventional Antibiotics ◽  
And Control

ABSTRACT Staphylococcus aureus is a major human pathogen, and the emergence of antibiotic-resistant strains is making all types of S. aureus infections more challenging to treat. With a pressing need to develop alternative control strategies to use alongside or in place of conventional antibiotics, one approach is the targeting of established virulence factors. However, attempts at this have had little success to date, suggesting that we need to better understand how this pathogen causes disease if effective targets are to be identified. To address this, using a functional genomics approach, we have identified a small membrane-bound protein that we have called MspA. Inactivation of this protein results in the loss of the ability of S. aureus to secrete cytolytic toxins, protect itself from several aspects of the human innate immune system, and control its iron homeostasis. These changes appear to be mediated through a change in the stability of the bacterial membrane as a consequence of iron toxicity. These pleiotropic effects on the ability of the pathogen to interact with its host result in significant impairment in the ability of S. aureus to cause infection in both a subcutaneous and sepsis model of infection. Given the scale of the effect the inactivation of MspA causes, it represents a unique and promising target for the development of a novel therapeutic approach.

Algorithm of real estate mass appraisal with inequality restricted least squares (IRLS) estimation

2020 ◽  
Vol 13 (2) ◽  
pp. 161-179
Author(s):  
Mariusz Doszyń
Keyword(s):  
Real Estate ◽  
Least Squares ◽  
Econometric Model ◽  
A Priori ◽  
Ordinary Least Squares ◽  
Real Estate Market ◽  
Negative Consequences ◽  
Content Type ◽  
Low Efficiency ◽  
The Impact

Purpose The purpose of this paper is to present an algorithm of real estate mass appraisal in which the impact of attributes (real estate features) is estimated by inequality restricted least squares (IRLS) model. Design/methodology/approach This paper presents the algorithm of real estate mass appraisal, which was also presented in the form of an econometric model. Vital problem related to econometric models of mass appraisal is multicollinearity. In this paper, a priori knowledge about parameters is used by imposing restrictions in the form of inequalities. IRLS model is therefore used to limit negative consequences of multicollinearity. In ordinary least squares (OLS) models, estimator variances might be inflated by multicollinearity, which could lead to wrong signs of estimates. In IRLS models, estimators efficiency is higher (estimator variances are lower), which could result in better appraisals. Findings The final effect of the analysis is a vector of the impact of real estate attributes on their value in the mass appraisal algorithm. After making expert corrections, the algorithm was used to evaluate 318 properties from the test set. Valuation errors were also discussed. Originality/value Restrictions in the form of inequalities were imposed on the parameters of the econometric model, ensuring the non-negativity and monotonicity of real estate attribute impact. In case of real estate, variables are usually correlated. OLS estimators are then inflated and inefficient. Imposing restrictions in form of inequalities could improve results because IRLS estimators are more efficient. In the case of results inconsistent with theoretical assumptions, the real estate mass appraisal algorithm enables having the obtained results adjusted by an expert. This can be important for low quality databases, which is often the case in underdeveloped real estate markets. Another reason for expert correction may be the low efficiency of a given real estate market.

Velocity-free distributed geometric formation control for underactuated UAVs

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Yanjie Chen ◽  
Weiwei Zhan ◽  
Yibin Huang ◽  
Zhiqiang Miao ◽  
Yaonan Wang
Keyword(s):  
Attitude Control ◽  
Formation Control ◽  
Control Method ◽  
Controller Design ◽  
Linear Velocity ◽  
Nonlinear Controller ◽  
Content Type ◽  
Distributed Controller ◽  
Cascade Structure ◽  
The Stability

Purpose This paper aims to investigate the distributed formation control problem for a multi-quadrotor unmanned aerial vehicle system without linear velocity feedbacks. Design/methodology/approach A nonlinear controller is proposed based on the orthogonal group SE(3) to obviate singularities and ambiguities of the traditional parameterized attitude representations. A cascade structure is applied in the distributed controller design. The inner loop is responsible for attitude control, and the outer loop is responsible for translational dynamics. To ensure a linear-velocity-free characteristic, some auxiliary variables are introduced to construct virtual signals in distributed controller design. The stability analysis of the proposed distributed control method by the Lyapunov function is provided as well. Findings A group of four quadrotors with constant reference linear velocity and a group of six quadrotors with varying reference linear velocity are adopted to verify the effectiveness of the proposed strategy. Originality/value This is a new innovation for multi-robot formation control method to improve assembly automation.

Guidance and control algorithms for mini UAV autopilots

2017 ◽  
Vol 89 (1) ◽  
pp. 133-144 ◽  
Author(s):  
Elisa Capello ◽  
Giorgio Guglieri ◽  
Gianluca Ristorto
Keyword(s):  
Attitude Control ◽  
Path Following ◽  
Adaptive Controller ◽  
Computationally Efficient ◽  
Control Laws ◽  
Content Type ◽  
Guidance And Control ◽  
Aerial Vehicle ◽  
Set Up ◽  
And Control

Purpose The aim of this paper is the implementation and validation of control and guidance algorithms for unmanned aerial vehicle (UAV) autopilots. Design/methodology/approach The path-following control of the UAV can be separated into different layers: inner loop for pitch and roll attitude control, outer loop on heading, altitude and airspeed control for the waypoints tracking and waypoint navigation. Two control laws are defined: one based on proportional integrative derivative (PID) controllers both for inner and outer loops and one based on the combination of PIDs and an adaptive controller. Findings Good results can be obtained in terms of trajectory tracking (based on waypoints) and of parameter variations. The adaptive control law guarantees smoothing responses and less oscillations and glitches on the control deflections. Practical implications The proposed controllers are easily implementable on-board and are computationally efficient. Originality/value The algorithm validation via hardware in the loop simulations can be used to reduce the platform set-up time and the risk of losing the prototype during the flight tests.

scholarly journals Evaluation and control process in higher education institutions: a comparative analysis

2019 ◽  
Vol 27 (3) ◽  
pp. 269-284
Author(s):  
Joaquim Mourato ◽  
Maria Teresa Patrício
Keyword(s):  
Higher Education ◽  
Performance Evaluation ◽  
Strategic Management ◽  
Performance Management ◽  
Binary Systems ◽  
Higher Education Institutions ◽  
Control Process ◽  
Content Type ◽  
Control Processes ◽  
And Control

Purpose The purpose of this paper is to study the evaluation and control processes in the governance systems of higher education institutions (HEIs). Design/methodology/approach This study compares the performance and control processes of strategic management in four HEIs in two European countries with binary systems – Portugal and The Netherlands. Using a case study approach, the authors find that HEIs with different missions and contexts have performance and control systems that are generally indistinguishable. Findings The controlling strategies in the public HEIs have taken on isomorphic characteristics based on processes that enhance competition, decentralize functions and solidify performance management. Originality/value In this paper, the authors “unpackage” strategic management to focus on the forms of control associated with performance evaluation. Performance evaluation is central to the management process and increasingly assuming an integral part of the institution’s identity and culture.

Design and control of a climbing robot for inspection of high mast lighting

2019 ◽  
Vol 39 (1) ◽  
pp. 77-85
Author(s):  
Peng Li ◽  
Xingguang Duan ◽  
Guangli Sun ◽  
Xiang Li ◽  
Yang Zhou ◽  
...  
Keyword(s):  
Tracking Error ◽  
Weight Ratio ◽  
Driving Mechanism ◽  
Climbing Robot ◽  
Adhesion Forces ◽  
Content Type ◽  
Compression Mechanism ◽  
The Stability ◽  
Wireless Module ◽  
And Control

Purpose This paper aims to develop a climbing robot to help people inspect lamps of high-mast lighting. Design/methodology/approach The robot consists of driving mechanism, suspension mechanism and compression mechanism. The driving mechanism is realized by link chains and sprockets, which are arranged opposite to each other, to form a dual caterpillar mechanism. The compression mechanism squeezes the caterpillar, and rubber feet “grasps” the steel rope to generate enough adhesion forces. The suspension mechanism is used to compensate the contraction or extension of the chains. The robot is equipped with a DC motor with a rated power of 250 W and a wireless module to communicate with the operator’s console. The dynamic model of the robot and the control strategy is derived, and the stability of the controller is proofed. Findings The payload experiment shows the robot can afford up to 3.7 times payload versus its own weight. Even when the payload is 30 kg, the robot can maintain a speed of the 1 m/s. The experiments also show that the tracking error of the robot reaches zero. Practical implications The proposed moving mechanism has a high load/weight ratio, which is a verified solution for the cable inspection purpose. Originality/value A rope climbing robot for high mast lighting inspection is proposed. The developed mechanism can reach a speed of 1 m/s with the payload of 30 kg, while its own weight is only 15.6 kg. The payload/weight ratio of the robot is 2.24; this value is rather good in many climbing robots reported in other renowned journal.

Gait planning and control of quadruped crawling robot on a slope

2019 ◽  
Vol 47 (1) ◽  
pp. 12-22
Author(s):  
Peng Wang ◽  
Chunxiao Song ◽  
Xiaoqiang Li ◽  
Peng Luo
Keyword(s):  
Position Control ◽  
Impedance Control ◽  
Switching Control ◽  
Gait Planning ◽  
Content Type ◽  
Planning And Control ◽  
Smooth Switching ◽  
The Stability ◽  
And Control ◽  
Soft Control

Purpose The gait planning and control of quadruped crawling robot affect the stability of the robot walking on a slope. The control includes the position control in the swing phase, the force control in the support phase and the switching control in the force/position switching. To improve the passing ability of quadruped crawling robot on a slope, this paper aims to propose a soft control strategy. Design/methodology/approach The strategy adopts the statically stable crawling gait as the main gait. As the robot moves forward, the position/force section switching control is adopted. When the foot does not touch the ground, the joint position control based on the variable speed PID is performed. When the foot touches the ground, the position-based impedance control is performed, and a fuzzy multi-model switching control based on friction compensation is proposed to achieve smooth switching of force and position. Findings The proposed method offers a solution for stable passage in slope environment. The quadruped crawling robot can realize smooth switching of force/position, precise positioning in the swing process and soft control of force in the supporting phase. This fact is verified by simulation and test. Originality/value The method presented in this paper takes advantage of minimal tracking errors and minimal jitters. Simulations and tests were performed to evaluate the performance.

Attitude manoeuvre control and asymptotic disturbance rejection of flexible spacecraft

2017 ◽  
Vol 233 (2) ◽  
pp. 426-437
Author(s):  
Zhen Yu ◽  
Yu Guo ◽  
Chenxing Zhong ◽  
Yifei Wu ◽  
Jian Guo
Keyword(s):  
Attitude Control ◽  
Disturbance Rejection ◽  
Coupling Effect ◽  
Flexible Spacecraft ◽  
Control Performance ◽  
Novel Approach ◽  
Attitude Controller ◽  
The Stability ◽  
Persistent Disturbances ◽  
Lyapunov Technique

This paper presents a novel approach to tackle the issues of attitude manoeuvre control and asymptotic disturbance rejection for a flexible spacecraft. The resulting attitude controller employs an internal model-based compensator to reject a class of persistent disturbances and a modal estimator for dynamic compensation of the rigid-flex coupling effect. The convergence of the modal variables can be guaranteed without the measures of them. The stability of the system is proved via the Lyapunov technique rigorously. Numerical results illustrate that improved attitude control performance and asymptotic disturbance rejection can both be achieved.

Sign in / Sign up

Export Citation Format

Share Document