SwingBot: Learning Physical Features from In-hand Tactile Exploration for Dynamic Swing-up Manipulation

• URL: http://arxiv.org/abs/2101.11812v1
• Date: Thu, 28 Jan 2021 04:35:15 GMT
• Title: SwingBot: Learning Physical Features from In-hand Tactile Exploration for Dynamic Swing-up Manipulation
• Authors: Chen Wang, Shaoxiong Wang, Branden Romero, Filipe Veiga, Edward Adelson
• Abstract summary: We present SwingBot, a robot that is able to learn the physical features of a held object through tactile exploration. We show that with the learned physical features our end-to-end self-supervised learning pipeline is able to substantially improve the accuracy of swinging up unseen objects.
• Score: 10.565804740298754
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Several robot manipulation tasks are extremely sensitive to variations of the physical properties of the manipulated objects. One such task is manipulating objects by using gravity or arm accelerations, increasing the importance of mass, center of mass, and friction information. We present SwingBot, a robot that is able to learn the physical features of a held object through tactile exploration. Two exploration actions (tilting and shaking) provide the tactile information used to create a physical feature embedding space. With this embedding, SwingBot is able to predict the swing angle achieved by a robot performing dynamic swing-up manipulations on a previously unseen object. Using these predictions, it is able to search for the optimal control parameters for a desired swing-up angle. We show that with the learned physical features our end-to-end self-supervised learning pipeline is able to substantially improve the accuracy of swinging up unseen objects. We also show that objects with similar dynamics are closer to each other on the embedding space and that the embedding can be disentangled into values of specific physical properties.

Related papers

• Learning Object Properties Using Robot Proprioception via Differentiable Robot-Object Interaction [52.12746368727368]
  Differentiable simulation has become a powerful tool for system identification. Our approach calibrates object properties by using information from the robot, without relying on data from the object itself. We demonstrate the effectiveness of our method on a low-cost robotic platform.
  arXiv  Detail & Related papers  (2024-10-04T20:48:38Z)
• RoboPack: Learning Tactile-Informed Dynamics Models for Dense Packing [38.97168020979433]
  We introduce an approach that combines visual and tactile sensing for robotic manipulation by learning a neural, tactile-informed dynamics model. Our proposed framework, RoboPack, employs a recurrent graph neural network to estimate object states. We demonstrate our approach on a real robot equipped with a compliant Soft-Bubble tactile sensor on non-prehensile manipulation and dense packing tasks.
  arXiv  Detail & Related papers  (2024-07-01T16:08:37Z)
• ManiFoundation Model for General-Purpose Robotic Manipulation of Contact Synthesis with Arbitrary Objects and Robots [24.035706461949715]
  There is a pressing need to develop a model that enables general-purpose robots to undertake a broad spectrum of manipulation tasks. Our work introduces a comprehensive framework to develop a foundation model for general robotic manipulation. Our model achieves average success rates of around 90%.
  arXiv  Detail & Related papers  (2024-05-11T09:18:37Z)
• Track2Act: Predicting Point Tracks from Internet Videos enables Generalizable Robot Manipulation [65.46610405509338]
  We seek to learn a generalizable goal-conditioned policy that enables zero-shot robot manipulation. Our framework,Track2Act predicts tracks of how points in an image should move in future time-steps based on a goal. We show that this approach of combining scalably learned track prediction with a residual policy enables diverse generalizable robot manipulation.
  arXiv  Detail & Related papers  (2024-05-02T17:56:55Z)
• DexTouch: Learning to Seek and Manipulate Objects with Tactile Dexterity [12.508332341279177]
  We introduce a multi-finger robot system designed to search for and manipulate objects using the sense of touch. To achieve this, binary tactile sensors are implemented on one side of the robot hand to minimize the Sim2Real gap. We demonstrate that object search and manipulation using tactile sensors is possible even in an environment without vision information.
  arXiv  Detail & Related papers  (2024-01-23T05:37:32Z)
• Synthesis and Execution of Communicative Robotic Movements with Generative Adversarial Networks [59.098560311521034]
  We focus on how to transfer on two different robotic platforms the same kinematics modulation that humans adopt when manipulating delicate objects. We choose to modulate the velocity profile adopted by the robots' end-effector, inspired by what humans do when transporting objects with different characteristics. We exploit a novel Generative Adversarial Network architecture, trained with human kinematics examples, to generalize over them and generate new and meaningful velocity profiles.
  arXiv  Detail & Related papers  (2022-03-29T15:03:05Z)
• From Movement Kinematics to Object Properties: Online Recognition of Human Carefulness [112.28757246103099]
  We show how a robot can infer online, from vision alone, whether or not the human partner is careful when moving an object. We demonstrated that a humanoid robot could perform this inference with high accuracy (up to 81.3%) even with a low-resolution camera. The prompt recognition of movement carefulness from observing the partner's action will allow robots to adapt their actions on the object to show the same degree of care as their human partners.
  arXiv  Detail & Related papers  (2021-09-01T16:03:13Z)
• Property-Aware Robot Object Manipulation: a Generative Approach [57.70237375696411]
  In this work, we focus on how to generate robot motion adapted to the hidden properties of the manipulated objects. We explore the possibility of leveraging Generative Adversarial Networks to synthesize new actions coherent with the properties of the object. Our results show that Generative Adversarial Nets can be a powerful tool for the generation of novel and meaningful transportation actions.
  arXiv  Detail & Related papers  (2021-06-08T14:15:36Z)
• Careful with That! Observation of Human Movements to Estimate Objects Properties [106.925705883949]
  We focus on the features of human motor actions that communicate insights on the weight of an object. Our final goal is to enable a robot to autonomously infer the degree of care required in object handling.
  arXiv  Detail & Related papers  (2021-03-02T08:14:56Z)
• Zero-shot sim-to-real transfer of tactile control policies for aggressive swing-up manipulation [5.027571997864706]
  This paper shows that robots equipped with a vision-based tactile sensor can perform dynamic manipulation tasks without prior knowledge of all the physical attributes of the objects to be manipulated. A robotic system is presented that is able to swing up poles of different masses, radii and lengths, to an angle of 180 degrees. This is the first work where a feedback policy from high-dimensional tactile observations is used to control the swing-up manipulation of poles in closed-loop.
  arXiv  Detail & Related papers  (2021-01-07T18:43:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.