Reinforcement learning-based shared control for walking-aid robot and its experimental verification

Xu, Wenxia; Huang, Jian; Wang, Yongji; Tao, Chunjing; Cheng, Lei

doi:10.1080/01691864.2015.1070748

Cited by 37 publications

(26 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The shared control is defined that a system can share its controller with one or more human beings and one or multiple robotic controllers [25]. In the field of shared control, many researchers gained quite a few achievements [26–30]. Overall, the research on the shared-control robot is still in its infancy.…”

Section: Introductionmentioning

confidence: 99%

Hierarchical Shared Control of Cane-Type Walking-Aid Robot

Tao

Yan

2017

Journal of Healthcare Engineering

Self Cite

View full text Add to dashboard Cite

A hierarchical shared-control method of the walking-aid robot for both human motion intention recognition and the obstacle emergency-avoidance method based on artificial potential field (APF) is proposed in this paper. The human motion intention is obtained from the interaction force measurements of the sensory system composed of 4 force-sensing registers (FSR) and a torque sensor. Meanwhile, a laser-range finder (LRF) forward is applied to detect the obstacles and try to guide the operator based on the repulsion force calculated by artificial potential field. An obstacle emergency-avoidance method which comprises different control strategies is also assumed according to the different states of obstacles or emergency cases. To ensure the user's safety, the hierarchical shared-control method combines the intention recognition method with the obstacle emergency-avoidance method based on the distance between the walking-aid robot and the obstacles. At last, experiments validate the effectiveness of the proposed hierarchical shared-control method.

show abstract

Section: Introductionmentioning

confidence: 99%

Hierarchical Shared Control of Cane-Type Walking-Aid Robot

Tao

Yan

2017

Journal of Healthcare Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, this study only used the TMAGRL for three classifications; the study about using TMAGRL for more classifications or continuous neural decoding can be explored. Additionally, learning from other RL system [ 52 ] or using a wearable sensor system [ 53 , 54 ] to determine the reward of the RL system may further improve the practicability. In the future, we intend to test our method integrated with a wireless wearable sensor system on a brain control task to further verify its effectiveness in clinical application.…”

Section: Discussionmentioning

confidence: 99%

Reinforcement Learning Based Fast Self-Recalibrating Decoder for Intracortical Brain–Machine Interface

Zhang

Chao

Chen

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Background: For the nonstationarity of neural recordings in intracortical brain–machine interfaces, daily retraining in a supervised manner is always required to maintain the performance of the decoder. This problem can be improved by using a reinforcement learning (RL) based self-recalibrating decoder. However, quickly exploring new knowledge while maintaining a good performance remains a challenge in RL-based decoders. Methods: To solve this problem, we proposed an attention-gated RL-based algorithm combining transfer learning, mini-batch, and weight updating schemes to accelerate the weight updating and avoid over-fitting. The proposed algorithm was tested on intracortical neural data recorded from two monkeys to decode their reaching positions and grasping gestures. Results: The decoding results showed that our proposed algorithm achieved an approximate 20% increase in classification accuracy compared to that obtained by the non-retrained classifier and even achieved better classification accuracy than the daily retraining classifier. Moreover, compared with a conventional RL method, our algorithm improved the accuracy by approximately 10% and the online weight updating speed by approximately 70 times. Conclusions: This paper proposed a self-recalibrating decoder which achieved a good and robust decoding performance with fast weight updating and might facilitate its application in wearable device and clinical practice.

show abstract

“…Sarsa-learning is widely used in the field of dynamic planning in an unexperienced environment [22][23]. In the study, Sarsa-learning is applied to calculate the optimal policy that is used to select optimal user control weight in different states.…”

Section: Choice Of the Online Reinforcement Learning Algorithmmentioning

confidence: 99%

Shared Control Design Methodologies of an Electric Wheelchair for Individuals with Severe Disabilities using Reinforcement Learning

Xi¹,

Shino²

2020

JASSE

View full text Add to dashboard Cite

Individuals with severe disabilities typically suffer from lack of muscle power and deformation of their hands. This significantly affects their ability to operate an Electric Wheelchair (EW) especially when using joystick controllers because one operation of a joystick affects both the straight and turning motions of an EW. Thus, individuals with severe disabilities typically exhibit difficulties in operating an EW, especially in the cases that require fine steering capabilities. In contrast to autonomous driving, the study proposes a novel shared control method using reinforcement learning considering utilizing users' residual physical functions. With the use of online learning, the shared controller provides proper assistance based on the driving environments and each user's input characteristics. The effectiveness and the characteristics of the novel shared controller are discussed via simulations.

show abstract

Reinforcement learning-based shared control for walking-aid robot and its experimental verification

Cited by 37 publications

References 27 publications

Hierarchical Shared Control of Cane-Type Walking-Aid Robot

Hierarchical Shared Control of Cane-Type Walking-Aid Robot

Reinforcement Learning Based Fast Self-Recalibrating Decoder for Intracortical Brain–Machine Interface

Shared Control Design Methodologies of an Electric Wheelchair for Individuals with Severe Disabilities using Reinforcement Learning

Contact Info

Product

Resources

About