Q-Learning-based model predictive variable impedance control for physical human-robot collaboration

Roveda, Loris; Testa, Andrea; Shahid, Asad Ali; Braghin, Francesco; Piga, Dario

doi:10.1016/j.artint.2022.103771

Cited by 32 publications

(18 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the human-robot collaborative motion experiment, after completing a point-to-point motion, each participant answered the questions in Table V according to the motion's naturalness, smoothness, stability, overall performance, etc. [12]. Questions 1-4 and 6 adopt 5-point Likert scale [36].…”

Section: Experiments Taskmentioning

confidence: 99%

“…Such as many papers are estimating the properties of the environment online to determine the control action guaranteeing stability or other performance [9,10]. Considering human-robot interaction, some papers are specifically estimating online the human-robot interaction dynamics to modulate the control action [11], guaranteeing the stability of the controller [12]. Therefore, the admittance control with fixed parameters is only applicable to the situation where the environment is fixed and the modeling is simple.To address the changing environment and improve the robot's interaction with the external environment, more researchers have studied variable admittance control of the robot.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Based on human-like variable admittance control for human–robot collaborative motion

Wang

Zhao

2023

Robotica

View full text Add to dashboard Cite

Admittance control of the robot is an important method to improve human–robot collaborative performance. However, it displays poor matching between admittance parameters and human–robot collaborative motion. This results in poor motion performance when the robot interacts with the changeable environment (human). Therefore, to improve the performance of human–robot collaboration, the human-like variable admittance parameter regulator (HVAPR) based on the change rate of interaction force is proposed by studying the human arm’s static and dynamic admittance parameters in human–human collaborative motion. HVAPR can generate admittance parameters matching with human collaborative motion. To test the performance of the proposed HVAPR, the human–robot collaborative motion experiment based on HVAPR is designed and compared with the variable admittance parameter regulator (VAPR). The satisfaction, recognition ratio, and recognition confidence of the two admittance parameter regulators are statistically analyzed via questionnaire. Simultaneously, the trajectory and interaction force of the robot are analyzed, and the performance of the human–robot collaborative motion is assessed and compared using the trajectory smoothness index and average energy index. The results show that HVAPR is superior to VAPR in human–robot collaborative satisfaction, robot trajectory smoothness, and average energy consumption.

show abstract

Section: Experiments Taskmentioning

confidence: 99%

mentioning

confidence: 99%

Based on human-like variable admittance control for human–robot collaborative motion

Wang

Zhao

2023

Robotica

View full text Add to dashboard Cite

show abstract

“…The problem of adapting the damping and stiffness terms is faced in [3], where Adversarial Inverse Reinforcement Learning (AIRL) is exploited to infer, starting from human "expert" demonstrations, the objective function against which the variable terms are optimized. Recent research has shown how Neural Networks (NNs) can be exploited to learn interaction dynamics [16], whether the contacts happens with a working environment [17] or a human being [18]. Estimating this dynamics can be exploited for the computation of optimal impedance gains for a safe and compliant interaction [19].…”

Section: B Related Workmentioning

confidence: 99%

“…In this scope, the methodology proposed in this article takes advantage of AI techniques to develop an optimizationbased control law. In literature, NNs learning the unknown interaction dynamics are mainly exploited for optimizing the robot compliance [17], [18], whereas our objective is delivering accurate force tracking. As regards the optimization of robot-environment interaction tasks, the state-of-art strategies suppose that an "expert" system is employed beforehand to register a policy, from which an optimal control configuration is devised [3], [15], with the policy being strongly taskdependent.…”

Section: B Related Workmentioning

confidence: 99%

“…Some design choices of the latter include its structural properties and the composition of the dataset used for its training. Similar NN models have been previously used for physical human-robot interaction [18]. Compared to them, we provide a straightforward setup that allows for autonomous learning of the interaction between the robot and an environment, which is of particular interest for the manipulators' deployment in industrial scenarios.…”

Section: Contributionmentioning

confidence: 99%

See 1 more Smart Citation

Optimized Residual Action for Interaction Control with Learned Environments

Puricelli¹,

Pozzi²,

Petrone³

et al. 2023

Preprint

View full text Add to dashboard Cite

Robotic tasks featuring interaction with other bodies are increasingly required in industrial contexts. The manipulators need to interact with the environment in a compliant way to avoid damage, but, at the same time, are often required to accurately track a reference force. To this aim, interaction controllers are typically employed, but they either need human tinkering for parameter tuning or precise modeling of the environment the robot will interact with. The former is a time-consuming procedure, while the latter is necessarily affected by approximations, which often lead to failure during the actual application. Both these aspects are problematic if it were often necessary to change the contact environment.Current research is concentrating on devising high-performance force controllers that are simple to tune and quick to adapt to changing environments. Along this line, this work proposes a novel control strategy, that we term ORACLE (Optimized Residual Action for interaction Control with Learned Environments). It exploits an ensemble of neural networks to estimate the force generated by the robot-environment interaction. This estimate is input to an optimal residual action controller that locally corrects the main action, output of a base force controller, which guarantees stability. The ORACLE strategy has been implemented and tested in the MuJoCo dynamic simulator and in a real-case scenario, both foreseeing a Franka Emika Panda robot used as a test platform. A reduction in terms of force tracking error is achieved by deploying the proposed strategy, with a short setup time.

show abstract

Human‐centered active torque‐based AAN impedance control for lower limb patient‐exoskeleton coupling system in the rehabilitation

Wang,

Guo,

Tao

et al. 2023

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

This article presents a human‐centered active torque‐based assist‐as‐needed (ATAAN) impedance control for lower limb patient‐exoskeleton coupling system (LLPECS), which considers the coupling effects (e.g., elastic connections, human soft tissues). The proposed double loop structure controller is composed of the establishment of the ATAAN impedance model in the outer‐loop and an adaptive interaction torque‐based controller in the inner‐loop. In the outer‐loop, the patient's muscle torque is estimated by a nonlinear disturbance observer, and the estimated result is used to approximate the patient's active torque. The parameters of ATAAN impedance model are modulated according to the approximation of active torque and the patient's position error to achieve the assist‐as‐needed property. In the inner‐loop, the desired interaction torque, which is required to realize the target ATAAN impedance model, is designed by a robust computed torque controller. The desired trajectory of the exoskeleton is planned online according to the desired interaction torque with the dynamics model of coupling effects. The prescribed performance controller is utilized to achieve accurate trajectory tracking of the exoskeleton, which further enables the interaction torque to track the desired value. The stability of the system and realization of the target ATAAN impedance model are rigorously proved by the Lyapunov method. Co‐simulation experiments are conducted to demonstrate the superiority of the proposed method compared with other existing controllers and its robustness against modeling uncertainties. The effectiveness of the proposed method has been verified through experiments in rehabilitation scenarios.

show abstract

Q-Learning-based model predictive variable impedance control for physical human-robot collaboration

Cited by 32 publications

References 69 publications

Based on human-like variable admittance control for human–robot collaborative motion

Based on human-like variable admittance control for human–robot collaborative motion

Optimized Residual Action for Interaction Control with Learned Environments

Human‐centered active torque‐based AAN impedance control for lower limb patient‐exoskeleton coupling system in the rehabilitation

Contact Info

Product

Resources

About