Multi-Modal Robot Apprenticeship: Imitation Learning Using Linearly Decayed DMP+ in a Human-Robot Dialogue System

Wu, Yan; Wang, Ruohan; D’Haro, Luis Fernando; Banchs, Rafael E.; Tee, Keng Peng

doi:10.1109/iros.2018.8593634

Cited by 15 publications

(20 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [38], a method called Compliant Parametric Dynamic Movement Primitives (CPDMP) has extended DMPs, enabling it to perform parametric learning on complex motion. In [39], [40], a method called DMP Plus has been proposed to achieve lower mean square error and efficient modification at the same time. The DMPs method has also been enhanced by reinforcement learning (RL) technique in achieving motion tuning [41].…”

Section: Related Workmentioning

confidence: 99%

A Framework of Hybrid Force/Motion Skills Learning for Robots

Wang

Chen

Nuovo

2021

IEEE Trans. Cogn. Dev. Syst.

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

A Framework of Hybrid Force/Motion Skills Learning for Robots

Wang

Chen

Nuovo

2021

IEEE Trans. Cogn. Dev. Syst.

View full text Add to dashboard Cite

“…These features are visualized in Fig. We compare our approach with several previous state of the art learning-based robot navigation techniques, including Learning from Demonstration (LfD) for robot navigation [14], multi-modal LfD (MM-LfD) [37], and Terrain Representation and Apprenticeship Learning (TRAL) [4]. To comprehensively evaluate the performance of robot navigation in a quantitative fashion, we use four evaluation metrics:…”

Section: A Experimental Setupsmentioning

confidence: 99%

“…SUCCESSFUL RUNS (WITH NO FAILURES) ARE USED TO CALCULATE THE METRICS OF TRAVERSAL TIME, INCONSISTENCY AND JERKINESS. OUR APPROACH IS COMPARED WITH LFD[14], MM-LFD[37] AND TRAL[4]. ) Terrain LfD MM-LfD TRAL Ours LfD MM-LfD TRAL Ours LfD MM-LfD TRAL Ours…”

mentioning

confidence: 99%

Robot Adaptation for Generating Consistent Navigational Behaviors over Unstructured Off-Road Terrain

Siva,

Wigness,

Rogers

et al. 2021

Preprint

View full text Add to dashboard Cite

Terrain adaptation is an essential capability for a ground robot to effectively traverse unstructured off-road terrain in real-world field environments such as forests. However, the expected robot behaviors generated by terrain adaptation methods cannot always be executed accurately due to setbacks such as wheel slip and reduced tire pressure. To address this problem, we propose a novel approach for consistent behavior generation that enables the ground robot's actual behaviors to more accurately match expected behaviors while adapting to a variety of unstructured off-road terrain. Our approach learns offset behaviors that are used to compensate for the inconsistency between the actual and expected behaviors without requiring the explicit modeling of various setbacks. Our approach is also able to estimate the importance of the multi-modal features to improve terrain representations for better adaptation. In addition, we develop an algorithmic solver for our formulated regularized optimization problem, which is guaranteed to converge to the global optimal solution. To evaluate the method, we perform extensive experiments using various unstructured off-road terrain in real-world field environments. Experimental results have validated that our approach enables robots to traverse complex unstructured off-road terrain with more navigational behavior consistency, and it outperforms previous methods, particularly so on challenging terrain.

show abstract

“…Lavine et al [ 18 ] also used deep neural network models to predict robot movements and achieved good results in robotic grabbing tasks based on hand-eye coordination. Abbeel et al [ 19 ] proposed an apprenticeship learning method, in which an expert attempts to maximize a reward function that is expressed as a linear combination of known features, and proposed an algorithm for learning a task demonstrated by an expert to give robots the ability to learn tasks for which they were not programmed [ 20 ]. Some researchers have combined DRL with cameras as the input of deep neural networks to develop robotic arm control.…”

Section: Related Workmentioning

confidence: 99%

A Multitasking-Oriented Robot Arm Motion Planning Scheme Based on Deep Reinforcement Learning and Twin Synchro-Control

Liu

Gao

et al. 2020

Sensors

View full text Add to dashboard Cite

Humanoid robots are equipped with humanoid arms to make them more acceptable to the general public. Humanoid robots are a great challenge in robotics. The concept of digital twin technology complies with the guiding ideology of not only Industry 4.0, but also Made in China 2025. This paper proposes a scheme that combines deep reinforcement learning (DRL) with digital twin technology for controlling humanoid robot arms. For rapid and stable motion planning for humanoid robots, multitasking-oriented training using the twin synchro-control (TSC) scheme with DRL is proposed. For switching between tasks, the robot arm training must be quick and diverse. In this work, an approach for obtaining a priori knowledge as input to DRL is developed and verified using simulations. Two simple examples are developed in a simulation environment. We developed a data acquisition system to generate angle data efficiently and automatically. These data are used to improve the reward function of the deep deterministic policy gradient (DDPG) and quickly train the robot for a task. The approach is applied to a model of the humanoid robot BHR-6, a humanoid robot with multiple-motion mode and a sophisticated mechanical structure. Using the policies trained in the simulations, the humanoid robot can perform tasks that are not possible to train with existing methods. The training is fast and allows the robot to perform multiple tasks. Our approach utilizes human joint angle data collected by the data acquisition system to solve the problem of a sparse reward in DRL for two simple tasks. A comparison with simulation results for controllers trained using the vanilla DDPG show that the designed controller developed using the DDPG with the TSC scheme have great advantages in terms of learning stability and convergence speed.

show abstract

Multi-Modal Robot Apprenticeship: Imitation Learning Using Linearly Decayed DMP+ in a Human-Robot Dialogue System

Cited by 15 publications

References 28 publications

A Framework of Hybrid Force/Motion Skills Learning for Robots

A Framework of Hybrid Force/Motion Skills Learning for Robots

Robot Adaptation for Generating Consistent Navigational Behaviors over Unstructured Off-Road Terrain

A Multitasking-Oriented Robot Arm Motion Planning Scheme Based on Deep Reinforcement Learning and Twin Synchro-Control

Contact Info

Product

Resources

About