Exploiting Abstract Symmetries in Reinforcement Learning for Complex Environments

Gupta, K. C.; Najjaran, Homayoun

doi:10.1109/icra46639.2022.9811652

Cited by 3 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meanwhile, the target networks are used to improve the stability of this approximation. Besides, the IBC-DMP agent is equipped with a dual-buffer structure, which is inspired by the previous work on off-policy RL [47], [58]. The demo buffer is used to store the demonstration data of the human motion recorded in Sec.…”

Section: A Overview Of the Training Methods For Ibc-dmp Rlmentioning

confidence: 99%

“…This method, however, renders complete separation between BC and agent training, such that BC is not helpful in improving the performance of RL. A recent study proposed a novel method to integrate BC into the training process of an RL agent, which greatly improves the convergence speed [34]. However, the demonstration used for BC is generated by a PID controller in a simulation environment, instead of real human data.…”

Section: *Corresponding Authormentioning

confidence: 99%

“…It also lacks the flexibility of achieving a balance between human likeness and the predefined reward. Recent studies made some attempts to integrate BC into the off-policy RL training using a dual-buffer structure [34], [58], which inspires us to use BC from human demonstration to leverage the training performance of RL.…”

Section: Behavior Cloning (Bc)mentioning

confidence: 99%

“…The form of the BC loss function ( 17) is inspired by the EBC-based RL in previous work which penalizes the deviation between the current policy π θ and the demonstration policy [34]. Nevertheless, (17) adopts the IBC technology which penalizes a certain energy function of the current policy [35].…”

Section: B Reshaped Actor Loss Based On Ibcmentioning

confidence: 99%

See 3 more Smart Citations

Motion generation for walking exoskeleton robot using multiple dynamic movement primitives sequences combined with reinforcement learning

Zhang

2022

Robotica

View full text Add to dashboard Cite

In order to assist patients with lower limb disabilities in normal walking, a new trajectory learning scheme of limb exoskeleton robot based on dynamic movement primitives (DMP) combined with reinforcement learning (RL) was proposed. The developed exoskeleton robot has six degrees of freedom (DOFs). The hip and knee of each artificial leg can provide two electric-powered DOFs for flexion/extension. And two passive-installed DOFs of the ankle were used to achieve the motion of inversion/eversion and plantarflexion/dorsiflexion. The five-point segmented gait planning strategy is proposed to generate gait trajectories. The gait Zero Moment Point stability margin is used as a parameter to construct a stability criteria to ensure the stability of human-exoskeleton system. Based on the segmented gait trajectory planning formation strategy, the multiple-DMP sequences were proposed to model the generation trajectories. Meanwhile, in order to eliminate the effect of uncertainties in joint space, the RL was adopted to learn the trajectories. The experiment demonstrated that the proposed scheme can effectively remove interferences and uncertainties.

show abstract

Section: A Overview Of the Training Methods For Ibc-dmp Rlmentioning

confidence: 99%

Section: *Corresponding Authormentioning

confidence: 99%

Section: Behavior Cloning (Bc)mentioning

confidence: 99%

Section: B Reshaped Actor Loss Based On Ibcmentioning

confidence: 99%

See 2 more Smart Citations

Motion generation for walking exoskeleton robot using multiple dynamic movement primitives sequences combined with reinforcement learning

Zhang

2022

Robotica

View full text Add to dashboard Cite

show abstract

“…Similarly to DL, RL, which is aimed at sequential decision-making, can be employed for motion planning in unfamiliar environments. It can resolve high-dimensional problems involving dynamic obstacles by taking into account their location over a limited number of timestamps within the past horizon [108].…”

Section: Reinforcement Learningmentioning

confidence: 99%

Conventional, Heuristic and Learning-Based Robot Motion Planning: Reviewing Frameworks of Current Practical Significance

2023

View full text Add to dashboard Cite

Motion planning algorithms have seen considerable progress and expansion across various domains of science and technology during the last few decades, where rapid advancements in path planning and trajectory optimization approaches have been made possible by the conspicuous enhancements brought, among others, by sampling-based methods and convex optimization strategies. Although they have been investigated from various perspectives in the existing literature, recent developments aimed at integrating robots into social, healthcare, industrial, and educational contexts have attributed greater importance to additional concepts that would allow them to communicate, cooperate, and collaborate with each other, as well as with human beings, in a meaningful and efficient manner. Therefore, in this survey, in addition to a brief overview of some of the essential aspects of motion planning algorithms, a few vital considerations required for assimilating robots into real-world applications, including certain instances of social, urban, and industrial environments, are introduced, followed by a critical discussion of a set of outstanding issues worthy of further investigation and development in future scientific studies.

show abstract

A review of recent trend in motion planning of industrial robots

Tamizi

Yaghoubi

Najjaran

2023

Int J Intell Robot Appl

View full text Add to dashboard Cite

Exploiting Abstract Symmetries in Reinforcement Learning for Complex Environments

Cited by 3 publications

References 10 publications

Motion generation for walking exoskeleton robot using multiple dynamic movement primitives sequences combined with reinforcement learning

Motion generation for walking exoskeleton robot using multiple dynamic movement primitives sequences combined with reinforcement learning

Conventional, Heuristic and Learning-Based Robot Motion Planning: Reviewing Frameworks of Current Practical Significance

A review of recent trend in motion planning of industrial robots

Contact Info

Product

Resources

About