Browsing by Subject "Tactile sensors"

Now showing 1 - 2 of 2

Open Access
Contact energy based hindsight experience prioritization
(IEEE, 2024-08-08) Sayar, Erdi; Bing, Zhenshan; D'Eramo, Carlo; Öğüz, Salih Özgür; Knoll, Alois
Multi-goal robot manipulation tasks with sparse rewards are difficult for reinforcement learning (RL) algorithms due to the inefficiency in collecting successful experiences. Recent algorithms such as Hindsight Experience Replay (HER) expedite learning by taking advantage of failed trajectories and replacing the desired goal with one of the achieved states so that any failed trajectory can be utilized as a contribution to learning. However, HER uniformly chooses failed trajectories, without taking into account which ones might be the most valuable for learning. In this paper, we address this problem and propose a novel approach Contact Energy Based Prioritization (CEBP) to select the samples from the replay buffer based on rich information due to contact, leveraging the touch sensors in the gripper of the robot and object displacement. Our prioritization scheme favors sampling of contact-rich experiences, which are arguably the ones providing the largest amount of information. We evaluate our proposed approach on various sparse reward robotic tasks and compare it with the state-of-the-art methods. We show that our method surpasses or performs on par with those methods on robot manipulation tasks. Finally, we deploy the trained policy from our method to a real Franka robot for a pick-and-place task. We observe that the robot can solve the task successfully. The videos and code are publicly available at: https://erdiphd.github.io/HER force/.
Open Access
Tactile perception by friction induced vibrations
(2011) Fagiani, R.; Massi, F.; Chatelet, E.; Berthier, Y.; Akay, A.
When a finger moves to scan the surface of an object (haptic sensing), the sliding contact generates vibrations that propagate in the finger skin activating the receptors (mechanoreceptors) located in the skin, allowing the brain to identify objects and perceive information about their properties. The information about the surface of the object is transmitted through vibrations induced by friction between the skin and the object scanned by the fingertip. The mechanoreceptors transduce the stress state into electrical impulses that are conveyed to the brain. A clear understanding of the mechanisms of the tactile sensing is fundamental to numerous applications, like the development of artificial tactile sensors for intelligent prostheses or robotic assistants, and in ergonomics. While the correlation between surface roughness and tactile sensation has already been reported in literature, the vibration spectra induced by the finger-surface scanning and the consequent activation of the mechanoreceptors on the skin have received less attention. In this paper, frequency analysis of signals characterizing surface scanning is carried out to investigate the vibration spectrum measured on the finger and to highlight the changes shown in the vibration spectra as a function of characteristic contact parameters such as scanning speed, roughness and surface texture. An experimental set-up is developed to recover the vibration dynamics by detecting the contact force and the induced vibrations; the bench test has been designed to guarantee reproducibility of measurements at the low amplitude of the vibrations of interest, and to perform measurements without introducing external noise. Two different perception mechanisms, as a function of the roughness wavelength, have been pointed out. The spectrum of vibration obtained by scanning textiles has been investigated. © 2011 Elsevier Ltd. All rights reserved.