Adaptive motion generation using imitation learning and highly compliant end effector for autonomous cleaning

@article{Ricardez2020AdaptiveMG,
  title={Adaptive motion generation using imitation learning and highly compliant end effector for autonomous cleaning},
  author={Gustavo Alfonso Garcia Ricardez and Nishanth Koganti and P.-C. Yang and Seigo Okada and Pedro Miguel Uriguen Eljuri and Akiya Yasuda and Lotfi El Hafi and M. Yamamoto and Jun Takamatsu and Tsukasa Ogasawara},
  journal={Advanced Robotics},
  year={2020},
  volume={34},
  pages={189 - 201}
}
Recent demographic trends in super aging societies, such as Japan, is leading to severe worker shortage. Service robots can play a promising role to augment human workers for performing various household and assistive tasks. Toilet cleanup is one such challenging task that involves performing complaint motion planning in a constrained toilet setting. In this study, we propose an end-to-end robotic framework to perform various tasks related to toilet cleanup. Our key contributions include the… 
View on Taylor & Francis
8 Citations
Background Citations
1
Methods Citations
1

8 Citations

Context Dependent Trajectory Generation using Sequence-to-Sequence Models for Robotic Toilet Cleaning
TLDR
A Sequence-to-Sequence (Seq2Seq) based neural network model is proposed to generate robot trajectories in configuration space given a context variable based on real-world measurements in Cartesian space using the offset between a target pose and the actual pose after localization as the context variable.
Development of Macro-Micro Robot with a Compliant End Effector for Putty Applying
TLDR
A 9 degree of freedom(DOF) redundant robot consisting of a 6 DOF series robot and a 3 DOF parallel robot is designed in this paper to achieve a flexible motion space and the feasibility of redundant robot kinematics algorithms and compliant EE is verified.
The Design of Stretch: A Compact, Lightweight Mobile Manipulator for Indoor Human Environments
TLDR
A novel design that reduces size, weight, and cost, while supporting a variety of tasks, is presented that provides empirical support from teleoperating and autonomously controlling a commercial robot based on the design to perform tasks in real homes.
Software development environment for collaborative research workflow in robotic system integration
TLDR
A containerized Software Development Environment (SDE) based on a collaborative workflow relying on open-source technologies for robotic system integration and deployment and shows that several research institutions successfully deployed it in multiple international competitions with various robotic platforms.
Motion Imitation for Monocular Videos
TLDR
The pose is used as an intermediate representation to perform this translation of a source video of a subject performing some movements or in motion, that movement/motion is transferred to amateur target in different motion.
Restock and straightening system for retail automation using compliant and mobile manipulation
TLDR
The proposed mobile manipulator features a custom-made end effector with compact and compliant design to safely and effectively manipulate products in retail stores to restock shelves, dispose expired products, and straighten products in Retail environments.
Verification experiment of stocking and disposal tasks by automatic shelf and mobile single-arm manipulation robot
A Visual Grasping Strategy for Improving Assembly Efficiency Based on Deep Reinforcement Learning
TLDR
A visual grasping strategy based on deep reinforcement learning is proposed for peg-in-hole assembly, which establishes a mapping relationship between the grasping position and the assembly time to improve the assembly efficiency.

References

SHOWING 1-10 OF 40 REFERENCES
Deep visual foresight for planning robot motion
  • Chelsea FinnS. Levine
  • Computer Science
    2017 IEEE International Conference on Robotics and Automation (ICRA)
  • 2017
TLDR
This work develops a method for combining deep action-conditioned video prediction models with model-predictive control that uses entirely unlabeled training data and enables a real robot to perform nonprehensile manipulation — pushing objects — and can handle novel objects not seen during training.
Task-Specific Generalization of Discrete and Periodic Dynamic Movement Primitives
TLDR
3-D vision on humanoid robots with complex oculomotor systems is often difficult due to the modeling uncertainties, but it is shown that these uncertainties can be accounted for by the proposed approach.
Floor cleaning robot with reconfigurable mechanism
Stereo-based real-time 6-DoF work tool tracking for robot programing by demonstration
TLDR
Tests show that the proposed system enables transferring to the machine the human ability of manipulating a spray gun, and is low-cost starting with a single pair of industrial cameras though more can be used for improved effectiveness and accuracy.
Movement reproduction and obstacle avoidance with dynamic movement primitives and potential fields
TLDR
A general framework for movement generation and mid-flight adaptation to obstacles is presented and obstacle avoidance is included by adding to the equations of motion a repellent force - a gradient of a potential field centered around the obstacle.
Autonomous Lavatory Cleaning System
The application of integrated robotics is becoming increasingly commonplace in day to day applications. The idea presented in this paper seeks to provide a convenient and a hassle-free means of…
Advanced tele-operated robot interface for teaching by demonstration
TLDR
The development of the advanced robotic interface integrated by electromyography (EMG) device, AR-10 Robotic hand, and Baxter robot is presented and the teaching by demonstration techniques has been presented using this integrated interface to perform two tasks.
Complex-YOLO: An Euler-Region-Proposal for Real-Time 3D Object Detection on Point Clouds
TLDR
Complex-YOLO, a state of the art real-time 3D object detection network on point clouds only, is introduced and a specific Euler-Region-Proposal Network (E-RPN) is proposed to estimate the pose of the object by adding an imaginary and a real fraction to the regression network.
RGB-D object detection and semantic segmentation for autonomous manipulation in clutter
TLDR
The experiments show that object detection and semantic segmentation complement each other and can be combined to yield reliable object perception.
Adaptive Coordination Strategies for Human-Robot Handovers
TLDR
This paper collected and analyzed data from human dyads performing a common household task—unloading a dish rack—and identified two coordination strategies that enabled givers to adapt to receivers’ task demands and formulated and implemented these strategies on a robotic manipulator.
...
...