Ring Gaussian Mixture Modelling and Regression for collaborative robots

Zaatari, Shirine El; Li, Weidong; Usman, Zahid

doi:10.1016/j.robot.2021.103864

Cited by 9 publications

(4 citation statements)

References 31 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…What's more, a method combining the advantages of DMP and ProMPs is also provided in Li et al (2023b), in which the ProMPs is embedded into neural networks to achieve efficient end-to-end learning of advanced trajectory statistics. The TP-GMM method, in Rozo et al (2015);El Zaatari et al (2021); Duque et al (2019) for instance, establishes a Gaussian mixture model by extracting probability models under the task and base coordinate systems, then reproduces trajectories through the Gaussian Mixture Regression (GMR). In this way, the method could learn the characteristics of the trajectory in specific task frame.…”

Section: Related Work Trajectory Learning Methods Based On Imitation...mentioning

confidence: 99%

Human skill knowledge guided global trajectory policy reinforcement learning method

Zang,

Wang,

Zha

et al. 2024

Front. Neurorobot.

View full text Add to dashboard Cite

Traditional trajectory learning methods based on Imitation Learning (IL) only learn the existing trajectory knowledge from human demonstration. In this way, it can not adapt the trajectory knowledge to the task environment by interacting with the environment and fine-tuning the policy. To address this problem, a global trajectory learning method which combinines IL with Reinforcement Learning (RL) to adapt the knowledge policy to the environment is proposed. In this paper, IL is proposed to acquire basic trajectory skills, and then learns the agent will explore and exploit more policy which is applicable to the current environment by RL. The basic trajectory skills include the knowledge policy and the time stage information in the whole task space to help learn the time series of the trajectory, and are used to guide the subsequent RL process. Notably, neural networks are not used to model the action policy and the Q value of RL during the RL process. Instead, they are sampled and updated in the whole task space and then transferred to the networks after the RL process through Behavior Cloning (BC) to get continuous and smooth global trajectory policy. The feasibility and the effectiveness of the method was validated in a custom Gym environment of a flower drawing task. And then, we executed the learned policy in the real-world robot drawing experiment.

show abstract

Section: Related Work Trajectory Learning Methods Based On Imitation...mentioning

confidence: 99%

Human skill knowledge guided global trajectory policy reinforcement learning method

Zang,

Wang,

Zha

et al. 2024

Front. Neurorobot.

View full text Add to dashboard Cite

show abstract

“…Moreover, a differentiation was made between the two types of frames, oriented and orientation-less, as defined by the authors [33]. An adjusted ring Gaussian model is used to model demonstration paths with respect to orientation-less frames [33].…”

Section: Training Tp-gmmmentioning

confidence: 99%

“…These relevant features are then used in TP-GMR to reproduce the new path. The ring Gaussians of orientation-less frames are transformed into normal Gaussians prior to performing TP-GMR [33]. In the calculation process of TP-GMR, (𝑥 𝑗 , 𝑦 𝑗 )is denoted as 𝑥 𝑗 .…”

Section: Calculating Tp-gmr For a Reproduced Pathmentioning

confidence: 99%

An improved approach of task-parameterized learning from demonstrations for cobots in dynamic manufacturing

Zaatari

Wang

et al. 2021

J Intell Manuf

View full text Add to dashboard Cite

show abstract

“…But, centralized data storage power is easily vulnerable to vicious attacks like single node cancellation and data interruption [11]. Many investigators are trying to use the newly formed block chain network technology in establishing an industrial block chain network system, to tackle the security problems of data transmission in industrial IoT [12]. Data approval between power notes is authorized by the Data Conference of the Smart Power Note [13].…”

Section: Introductionmentioning

confidence: 99%

Theil-Sen Regressive Miyaguchi–Preneel Based Cryptographic Hash Blockchain for Secure Data Transmission Using Remote Sensing Data in Iot

Sharmila¹,

2022

SSRN Journal

View full text Add to dashboard Cite

Ring Gaussian Mixture Modelling and Regression for collaborative robots

Cited by 9 publications

References 31 publications

Human skill knowledge guided global trajectory policy reinforcement learning method

Human skill knowledge guided global trajectory policy reinforcement learning method

An improved approach of task-parameterized learning from demonstrations for cobots in dynamic manufacturing

Theil-Sen Regressive Miyaguchi–Preneel Based Cryptographic Hash Blockchain for Secure Data Transmission Using Remote Sensing Data in Iot

Contact Info

Product

Resources

About