Confidence-Based Multi-Robot Learning from Demonstration

@article{Chernova2010ConfidenceBasedML,
  title={Confidence-Based Multi-Robot Learning from Demonstration},
  author={S. Chernova and Manuela M. Veloso},
  journal={International Journal of Social Robotics},
  year={2010},
  volume={2},
  pages={195-215}
}
Learning from demonstration algorithms enable a robot to learn a new policy based on demonstrations provided by a teacher. In this article, we explore a novel research direction, multi-robot learning from demonstration, which extends demonstration based learning methods to collaborative multi-robot domains. Specifically, we study the problem of enabling a single person to teach individual policies to multiple robots at the same time. We present flexMLfD, a task and platform independent multi… 
A framework for learning by demonstration in multi-teacher multi-robot scenarios
TLDR
The core contribution of this thesis is the design and implementation of a novel, multilayered framework for multi-robot learning from simultaneous demonstrations, capable of deriving control policies at two different levels of abstraction.
Learning from Demonstration for Distributed, Encapsulated Evolution of Autonomous Outdoor Robots
TLDR
These experiments compare robotic performance on a task after distributing the behaviors or the simple demonstrations to performance using a non-distributed LfD model receiving complex demonstrations, and show a strong improvement to behavior when using distributed simple demonstrations.
Perturbation Training for Human-Robot Teams
TLDR
Results from large-scale human subject experiments indicate that AdaPT enables an agent to learn in a manner compatible with a human's own learning process, and that a robot undergoing perturbation training with ahuman results in a high level of team performance.
Perturbation Training for Human-Robot Teams by Ramya Ramakrishnan
TLDR
This work designs and evaluates a computational learning model that enables a human-robot team to co-develop joint strategies for performing novel tasks requiring coordination and augments this model with a co-learning framework and a computational bi-directional communication protocol so that the robot can work with a person in live interactions.
Team Learning from Demonstration (TLfD): A Framework to Build Collaboration in a Team of Agents via Imitation
TLDR
This dissertation presents a framework of Team Learning from Demonstration (TLfD) that allows a group of human experts to train a team of agents via demonstrations and shows the effectiveness of TLfD framework and supported the idea of using imitation to build collaboration among multiple agents.
Towards improving incremental learning of manipulator kinematics with inter-robot knowledge transfer
  • Ndivhuwo Makondo, Benjamin Rosman
  • Computer Science
    2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)
  • 2019
TLDR
This paper explores knowledge transfer in the context of learning arm kinematics models, where an experienced robot shares its kinematic data with a new robot that is autonomously exploring its environment and shows that the sensorimotor models of the new robot can be bootstrapped by the shared knowledge, converge faster and achieve a better asymptotic performance compared to individual exploration from scratch.
Teaching multiple robots by a human — Teaching data generation
TLDR
A service system composed of small robots able to do tasks executed by a single human in environments like homes or offices and using kinematic data during the teaching process to determine how many robots will execute the task and how to split it into subtasks for each robot.
A Non-Linear Manifold Alignment Approach to Robot Learning from Demonstrations
TLDR
This paper proposes a method that enables robots with unknown kinematics to learn skills from demonstrations, which requires a motion trajectory obtained from human demonstrations via a vision-based system, which is then projected onto a corresponding human skeletal model.
Training a Single Agent with HiTAB
TLDR
This paper presents and compares two approaches to training teams of heterogenous robots: first, forming a multiagent control hierarchy which scales to large numbers of robots but requires the training of additional virtual controller agents; and second, modifying each robot’s feature space to include information about other robots’ current behaviors or limited internal state.
...
1
2
3
4
...

References

SHOWING 1-10 OF 63 REFERENCES
Teaching multi-robot coordination using demonstration of communication and state sharing
TLDR
This work introduces teaching by demonstration in the context of multi-robot tasks, enabling a single teacher to instruct multiple robots to work together through a demonstration of the desired behavior.
Multi-thresholded approach to demonstration selection for interactive robot learning
  • S. Chernova, M. Veloso
  • Computer Science
    2008 3rd ACM/IEEE International Conference on Human-Robot Interaction (HRI)
  • 2008
TLDR
This paper focuses on a demonstration selection algorithm used by the robot to identify informative states for demonstration, and presents a comparison of its multi-threshold selection method to confidence-based selection using a single fixed threshold, and to manual data selection by a human teacher.
Confidence-based robot policy learning from demonstration
TLDR
This thesis introduces Confidence-Based Autonomy, a mixed-initiative single robot demonstration learning algorithm that enables the robot and teacher to jointly control the learning process and selection of demonstration training data.
Robot Learning From Demonstration
TLDR
This work has shown that incorporating a task level direct learning component, which is non-model-based, in addition to the model-based planner, is useful in compensating for structural modeling errors and slow model learning.
Interactive Policy Learning through Confidence-Based Autonomy
TLDR
The algorithm selects demonstrations based on a measure of action selection confidence, and results show that using Confident Execution the agent requires fewer demonstrations to learn the policy than when demonstrations are selected by a human teacher.
A survey of robot learning from demonstration
Reinforcement learning for imitating constrained reaching movements
TLDR
A system to teach the robot constrained reaching tasks is described based on a dynamic system generator modulated by a learned speed trajectory combined with a reinforcement learning module to allow the robot to adapt the trajectory when facing a new situation, e.g., in the presence of obstacles.
Dogged Learning for Robots
  • D. Grollman, O. Jenkins
  • Computer Science
    Proceedings 2007 IEEE International Conference on Robotics and Automation
  • 2007
TLDR
A learning system (dogged learning) that combines learning from demonstration and mixed initiative control to enable lifelong learning for unknown tasks is presented.
Tutelage and socially guided robot learning
  • A. Thomaz, C. Breazeal
  • Computer Science
    2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566)
  • 2004
TLDR
A learning mechanism is presented, implemented on a humanoid robot, to demonstrate that a collaborative dialog framework allows a robot to efficiently learn a task from a human, generalize this ability to a new task configuration, and show commitment to the overall goal of the learned task.
Teaching and working with robots as a collaboration
  • C. Breazeal, Guy Hoffman, A. Thomaz
  • Computer Science
    Proceedings of the Third International Joint Conference on Autonomous Agents and Multiagent Systems, 2004. AAMAS 2004.
  • 2004
TLDR
This work uses collaborative discourse with accompanying gestures and social cues to teach a humanoid robot a structurally complex task and dynamically mesh its plans with those of its partner, according to the relative capabilities of the human and the robot.
...
1
2
3
4
5
...