Adaptive Control of Robotic Manipulators With Unified Motion Constraints

Li, Mingming; Li, Yanan; Ge, Shuzhi Sam; Lee, Tong Heng

doi:10.1109/tsmc.2016.2608969

Cited by 60 publications

(28 citation statements)

References 51 publications

(62 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where y m is the output of reference model, K is the parameters vector and e is the signal error. To make the system robust against external disturbances and uncertainties, term u c can be added to the control law (12) as follow which is the H ∞ control term.…”

Section: Second-type Adaptive Fuzzy Controller Designmentioning

confidence: 99%

“…Here, control and acceleration constraints are combined and converted into common input constraints. In [12], an adaptive-neural control is developed to obtain tracking when the issue of displacement constraints is addressed by considering the normal input saturation effect. The bound in all closed loop signals is monotone through Lyapunov stability analysis.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

One DOF Robot Manipulator Control Through Type-2 Fuzzy Robust Adaptive Controller

Naderolasli

Chatraei

2019

JAMRIS

View full text Add to dashboard Cite

In this article, the one DOF robot manipulator control is assessed through second type robust fuzzy-adaptive controller. The objective is to obtain a tracking path with appropriate accuracy. The stability of the closed loop system is verified through Lyapunov stability theory and the efficiency of tracking is analyzed subject to the constraints and uncertainty. In order to design the fuzzy controller a set of if-then fuzzy rules are considered which describe the system input-output behavior. Simulation and the results of the experiments on the one DOF robots indicate the effectiveness of the proposed methods.

show abstract

Section: Second-type Adaptive Fuzzy Controller Designmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

One DOF Robot Manipulator Control Through Type-2 Fuzzy Robust Adaptive Controller

Naderolasli

Chatraei

2019

JAMRIS

View full text Add to dashboard Cite

show abstract

“…is the position vector of brake i with x (i) r and y (i) r being the x r and y r components, respectively, andk is a unit vector perpendicular to the x − y plane. Using the nonholonomic constraint relation given by (12), (14) simplifies to…”

Section: Kineticsmentioning

confidence: 99%

Parking Control of Mixed Conventional/Braking Actuation Mobile Robots Using Fuzzy Logic Control

Nikshi

Bedillion

Hoover

2016

Volume 4A: Dynamics, Vibration, and Control

View full text Add to dashboard Cite

The use of conventional actuators in robotic systems (electric motors in particular), while often offering advantages in terms of flexibility and controllability, suffer from primary actuator failure, due to unexpected complexities in their environment, which can lead to loss of controllability. Conventional actuators can impose disadvantages on mechanical complexity, weight, and cost. Here, the Mixed conventional/braking Actuation Mobile Robot (MAMR), a new mobile robot platform, is proposed to tackle such drawbacks in actuation and explore the use and control of braking actuation. This platform substitutes the drive motors used in Ackermann steering with brakes that have only two states, ON and OFF. Additionally, the conventional drive wheels are replaced by a single, omni-directional wheel that only supports a driving force in the robots longitudinal direction. The ability of braking actuators in providing controllability under actuator failure is one of the primary motivations of this work. To validate the reliability and accuracy of MAMR approach, this paper studies the design of such robotic systems, the design and synthesis of fuzzy logic controllers along with the experimental assessments of these controllers in real-time. The experimental tests point out the controller performance enhancement using fuzzy logic controllers and MAMR.

show abstract

“…This enhanced its generalization capability, by increasing the diversity of training data. Nevertheless, current GPS schemes can only train policies with a batch mode for different tasks, and are known to struggle with challenges of incremental data processing, particularly in robotic applications [12], [13], [14]. Specifically, GPS methods will not work if all training tasks are presented sequentially, and not collectively made available during the early training period.…”

Section: Introductionmentioning

confidence: 99%

Guided Policy Search for Sequential Multitask Learning

Xiong

Sun

Yang

et al. 2019

IEEE Trans. Syst. Man Cybern, Syst.

View full text Add to dashboard Cite

Abstract-Policy search in reinforcement learning (RL) is a practical approach to interact directly with environments in parameter spaces, that often deal with dilemmas of local optima and real-time sample collection. A promising algorithm, known as guided policy search (GPS), is capable of handling the challenge of training samples using trajectory-centric methods. It can also provide asymptotic local convergence guarantees. However, in its current form, the GPS algorithm cannot operate in sequential multi-task learning scenarios. This is due to its batch-style training requirement, where all training samples are collectively provided at the start of the learning process. The algorithm's adaptation is thus hindered for real-time applications, where training samples or tasks can arrive randomly. In this paper, the GPS approach is reformulated, by adapting a recently proposed, lifelong-learning method, elastic weight consolidation (EWC). Specifically, Fisher information is incorporated to impart knowledge from previously learned tasks. The proposed algorithm, termed sequential multi-task, learning-guided policy search (SMT-GPS), is able to operate in sequential multi-task learning settings, ensuring continuous policy learning, without catastrophic forgetting. Pendulum and robotic manipulation experiments demonstrate the new algorithms efficacy to learn control policies for handling sequentially-arriving training samples, delivering comparable performance to the traditional, batch-based GPS algorithm. In conclusion, the proposed algorithm is posited as a new benchmark for the real-time RL and robotics research community.Index Terms-Reinforcement learning, guided policy search, sequential multi-task learning, elastic weight consolidation.

show abstract

Adaptive Control of Robotic Manipulators With Unified Motion Constraints

Cited by 60 publications

References 51 publications

One DOF Robot Manipulator Control Through Type-2 Fuzzy Robust Adaptive Controller

One DOF Robot Manipulator Control Through Type-2 Fuzzy Robust Adaptive Controller

Parking Control of Mixed Conventional/Braking Actuation Mobile Robots Using Fuzzy Logic Control

Guided Policy Search for Sequential Multitask Learning

Contact Info

Product

Resources

About