-
Learning Skateboarding for Humanoid Robots through Massively Parallel Reinforcement Learning
Authors:
William Thibault,
Vidyasagar Rajendran,
William Melek,
Katja Mombaur
Abstract:
Learning-based methods have proven useful at generating complex motions for robots, including humanoids. Reinforcement learning (RL) has been used to learn locomotion policies, some of which leverage a periodic reward formulation. This work extends the periodic reward formulation of locomotion to skateboarding for the REEM-C robot. Brax/MJX is used to implement the RL problem to achieve fast train…
▽ More
Learning-based methods have proven useful at generating complex motions for robots, including humanoids. Reinforcement learning (RL) has been used to learn locomotion policies, some of which leverage a periodic reward formulation. This work extends the periodic reward formulation of locomotion to skateboarding for the REEM-C robot. Brax/MJX is used to implement the RL problem to achieve fast training. Initial results in simulation are presented with hardware experiments in progress.
△ Less
Submitted 12 September, 2024;
originally announced September 2024.
-
Droplets in Acoustic Fields: A Unified Theory from Migration to Splitting
Authors:
Jeyapradhap Thirisangu,
Varun Kumar Rajendran,
Snekan Selvakannan,
Sujith Jayakumar,
E Hemachandran,
Karthick Subramani
Abstract:
We present a comprehensive theoretical framework governing the dynamics of droplets in acoustic fields, applicable to all droplet sizes, from the Rayleigh limit (D<<lambda) and beyond. Our theory elucidates the nature of acoustic forces and incorporates the effects of interfacial tension, enabling predictions of droplet migration, deformation, and splitting under suitable conditions. Importantly,…
▽ More
We present a comprehensive theoretical framework governing the dynamics of droplets in acoustic fields, applicable to all droplet sizes, from the Rayleigh limit (D<<lambda) and beyond. Our theory elucidates the nature of acoustic forces and incorporates the effects of interfacial tension, enabling predictions of droplet migration, deformation, and splitting under suitable conditions. Importantly, we demonstrate droplet deformation and splitting through bulk acoustic wave (BAW) silicon-glass microscale experiments, which validate the proposed theory.
△ Less
Submitted 12 August, 2024;
originally announced August 2024.
-
Enabling Tactile Feedback for Robotic Strawberry Handling using AST Skin
Authors:
Vishnu Rajendran,
Kiyanoush Nazari,
Simon Parsons,
Amir Ghalamzan
Abstract:
Acoustic Soft Tactile (AST) skin is a novel sensing technology which derives tactile information from the modulation of acoustic waves travelling through the skin's embedded acoustic channels. A generalisable data-driven calibration model maps the acoustic modulations to the corresponding tactile information in the form of contact forces with their contact locations and contact geometries. AST ski…
▽ More
Acoustic Soft Tactile (AST) skin is a novel sensing technology which derives tactile information from the modulation of acoustic waves travelling through the skin's embedded acoustic channels. A generalisable data-driven calibration model maps the acoustic modulations to the corresponding tactile information in the form of contact forces with their contact locations and contact geometries. AST skin technology has been highlighted for its easy customisation. As a case study, this paper discusses the possibility of using AST skin on a custom-built robotic end effector finger for strawberry handling. The paper delves into the design, prototyping, and calibration method to sensorise the end effector finger with AST skin. A real-time force-controlled gripping experiment is conducted with the sensorised finger to handle strawberries by their peduncle. The finger could successfully grip the strawberry peduncle by maintaining a preset force of 2 N with a maximum Mean Absolute Error (MAE) of 0.31 N over multiple peduncle diameters and strawberry weight classes. Moreover, this study sets confidence in the usability of AST skin in generating real-time tactile feedback for robot manipulation tasks.
△ Less
Submitted 1 July, 2024;
originally announced July 2024.
-
Single and bi-layered 2-D acoustic soft tactile skin (AST2)
Authors:
Vishnu Rajendran,
Simon Parsons,
Amir Ghalamzan E
Abstract:
This paper aims to present an innovative and cost-effective design for Acoustic Soft Tactile (AST) Skin, with the primary goal of significantly enhancing the accuracy of 2-D tactile feature estimation. The existing challenge lies in achieving precise tactile feature estimation, especially concerning contact geometry characteristics, using cost-effective solutions. We hypothesise that by harnessing…
▽ More
This paper aims to present an innovative and cost-effective design for Acoustic Soft Tactile (AST) Skin, with the primary goal of significantly enhancing the accuracy of 2-D tactile feature estimation. The existing challenge lies in achieving precise tactile feature estimation, especially concerning contact geometry characteristics, using cost-effective solutions. We hypothesise that by harnessing acoustic energy through dedicated acoustic channels in 2 layers beneath the sensing surface and analysing amplitude modulation, we can effectively decode interactions on the sensory surface, thereby improving tactile feature estimation. Our approach involves the distinct separation of hardware components responsible for emitting and receiving acoustic signals, resulting in a modular and highly customizable skin design. Practical tests demonstrate the effectiveness of this novel design, achieving remarkable precision in estimating contact normal forces (MAE < 0.8 N), 2D contact localisation (MAE < 0.7 mm), and contact surface diameter (MAE < 0.3 mm). In conclusion, the AST skin, with its innovative design and modular architecture, successfully addresses the challenge of tactile feature estimation. The presented results showcase its ability to precisely estimate various tactile features, making it a practical and cost-effective solution for robotic applications.
△ Less
Submitted 29 February, 2024; v1 submitted 25 January, 2024;
originally announced January 2024.
-
Deep Functional Predictive Control for Strawberry Cluster Manipulation using Tactile Prediction
Authors:
Kiyanoush Nazari,
Gabriele Gandolfi,
Zeynab Talebpour,
Vishnu Rajendran,
Paolo Rocco,
Amir Ghalamzan E.
Abstract:
This paper introduces a novel approach to address the problem of Physical Robot Interaction (PRI) during robot pushing tasks. The approach uses a data-driven forward model based on tactile predictions to inform the controller about potential future movements of the object being pushed, such as a strawberry stem, using a robot tactile finger. The model is integrated into a Deep Functional Predictiv…
▽ More
This paper introduces a novel approach to address the problem of Physical Robot Interaction (PRI) during robot pushing tasks. The approach uses a data-driven forward model based on tactile predictions to inform the controller about potential future movements of the object being pushed, such as a strawberry stem, using a robot tactile finger. The model is integrated into a Deep Functional Predictive Control (d-FPC) system to control the displacement of the stem on the tactile finger during pushes. Pushing an object with a robot finger along a desired trajectory in 3D is a highly nonlinear and complex physical robot interaction, especially when the object is not stably grasped. The proposed approach controls the stem movements on the tactile finger in a prediction horizon. The effectiveness of the proposed FPC is demonstrated in a series of tests involving a real robot pushing a strawberry in a cluster. The results indicate that the d-FPC controller can successfully control PRI in robotic manipulation tasks beyond the handling of strawberries. The proposed approach offers a promising direction for addressing the challenging PRI problem in robotic manipulation tasks. Future work will explore the generalisation of the approach to other objects and tasks.
△ Less
Submitted 9 March, 2023;
originally announced March 2023.
-
Target Domain Data induces Negative Transfer in Mixed Domain Training with Disjoint Classes
Authors:
Eryk Banatt,
Vickram Rajendran,
Liam Packer
Abstract:
In practical scenarios, it is often the case that the available training data within the target domain only exist for a limited number of classes, with the remaining classes only available within surrogate domains. We show that including the target domain in training when there exist disjoint classes between the target and surrogate domains creates significant negative transfer, and causes perform…
▽ More
In practical scenarios, it is often the case that the available training data within the target domain only exist for a limited number of classes, with the remaining classes only available within surrogate domains. We show that including the target domain in training when there exist disjoint classes between the target and surrogate domains creates significant negative transfer, and causes performance to significantly decrease compared to training without the target domain at all. We hypothesize that this negative transfer is due to an intermediate shortcut that only occurs when multiple source domains are present, and provide experimental evidence that this may be the case. We show that this phenomena occurs on over 25 distinct domain shifts, both synthetic and real, and in many cases deteriorates the performance to well worse than random, even when using state-of-the-art domain adaptation methods.
△ Less
Submitted 2 March, 2023;
originally announced March 2023.
-
On the stability of inhomogeneous fluids under acoustic fields
Authors:
Varun Kumar Rajendran,
Aravind Ram S P,
Karthick Subramani
Abstract:
In this work, we present the stability theory for inhomogeneous fluids subjected to standing acoustic fields. Starting from the first principles, the stability criterion is established for two fluids of different acoustic impedance separated by a plane interface. Through stability theory and numerical simulations we show that, in the presence of interfacial tension, the relocation of high-impedanc…
▽ More
In this work, we present the stability theory for inhomogeneous fluids subjected to standing acoustic fields. Starting from the first principles, the stability criterion is established for two fluids of different acoustic impedance separated by a plane interface. Through stability theory and numerical simulations we show that, in the presence of interfacial tension, the relocation of high-impedance fluid from anti-node to node occurs when the acoustic force overcomes interfacial tension force, which is in agreement with recent microchannel experiments. Furthermore, we establish an acoustic Bond number that characterizes stable (Bo_a < 1) and relocation (Bo_a > 1) regimes. Remarkably, it is found that the critical acoustic energy density required for relocation can be significantly reduced by increasing the channel height which could help design acoustofluidic microchannel devices that handle immiscible fluids.
△ Less
Submitted 8 December, 2022;
originally announced December 2022.
-
Shape-Biased Domain Generalization via Shock Graph Embeddings
Authors:
Maruthi Narayanan,
Vickram Rajendran,
Benjamin Kimia
Abstract:
There is an emerging sense that the vulnerability of Image Convolutional Neural Networks (CNN), i.e., sensitivity to image corruptions, perturbations, and adversarial attacks, is connected with Texture Bias. This relative lack of Shape Bias is also responsible for poor performance in Domain Generalization (DG). The inclusion of a role of shape alleviates these vulnerabilities and some approaches h…
▽ More
There is an emerging sense that the vulnerability of Image Convolutional Neural Networks (CNN), i.e., sensitivity to image corruptions, perturbations, and adversarial attacks, is connected with Texture Bias. This relative lack of Shape Bias is also responsible for poor performance in Domain Generalization (DG). The inclusion of a role of shape alleviates these vulnerabilities and some approaches have achieved this by training on negative images, images endowed with edge maps, or images with conflicting shape and texture information. This paper advocates an explicit and complete representation of shape using a classical computer vision approach, namely, representing the shape content of an image with the shock graph of its contour map. The resulting graph and its descriptor is a complete representation of contour content and is classified using recent Graph Neural Network (GNN) methods. The experimental results on three domain shift datasets, Colored MNIST, PACS, and VLCS demonstrate that even without using appearance the shape-based approach exceeds classical Image CNN based methods in domain generalization.
△ Less
Submitted 12 September, 2021;
originally announced September 2021.
-
Theory of Nonlinear Acoustic Forces Acting on Inhomogeneous Fluids
Authors:
Varun Kumar Rajendran,
Sujith Jayakumar,
Mohammed Azharudeen,
Karthick Subramani
Abstract:
Recently, the phenomena of streaming suppression and relocation of inhomogeneous miscible fluids under acoustic fields were explained using the hypothesis on mean Eulerian pressure. In this letter, we show that this hypothesis is unsound and any assumption on mean Eulerian pressure is needless. We present a theory of non-linear acoustics for inhomogeneous fluids from the first principles, which ex…
▽ More
Recently, the phenomena of streaming suppression and relocation of inhomogeneous miscible fluids under acoustic fields were explained using the hypothesis on mean Eulerian pressure. In this letter, we show that this hypothesis is unsound and any assumption on mean Eulerian pressure is needless. We present a theory of non-linear acoustics for inhomogeneous fluids from the first principles, which explains streaming suppression and acoustic relocation in both miscible and immiscible inhomogeneous fluids inside a microchannel. This theory predicts the relocation of higher impedance fluids to pressure nodes of the standing wave, which agrees with the recent experiments.
△ Less
Submitted 26 August, 2021;
originally announced August 2021.
-
SCUBA: An In-Device Multiplexed Protocol for Sidelink Communication on Unlicensed Bands
Authors:
Vishnu Rajendran,
Gautham Prasad,
Lutz Lampe,
Gus Vos
Abstract:
Device-to-device communication (D2D) is a key enabler for connecting devices together to form the Internet of Things (IoT). A growing issue with IoT networks is the increasing number of IoT devices congesting the spectral resources of the cellular bands. Operating D2D in unlicensed band alleviates this issue by offloading network traffic from the licensed bands, while also reducing the associated…
▽ More
Device-to-device communication (D2D) is a key enabler for connecting devices together to form the Internet of Things (IoT). A growing issue with IoT networks is the increasing number of IoT devices congesting the spectral resources of the cellular bands. Operating D2D in unlicensed band alleviates this issue by offloading network traffic from the licensed bands, while also reducing the associated licensing costs. To this end, we present a new low-cost radio access technology (RAT) protocol, called Sidelink Communications on Unlicensed BAnds (SCUBA), which can be implemented on cellular devices such that it coexists with the legacy cellular protocol by operating as a secondary RAT in a time division duplex manner using the existing radio hardware. SCUBA is compatible on different types of cellular devices including the low-complexity half-duplex frequency division duplex machine type communication (MTC) user equipments. SCUBA provides flexible sidelink (SL) latency and battery life tradeoff using a discontinuous reception procedure, which ensures that it is applicable across a wide range of use cases. We prove the effectiveness of our protocol with analyses and simulation results of the medium access control layer of SCUBA using different types of MTC traffic for both SL and the underlying cellular communication.
△ Less
Submitted 2 March, 2021; v1 submitted 7 December, 2020;
originally announced December 2020.
-
Accurate Layerwise Interpretable Competence Estimation
Authors:
Vickram Rajendran,
William LeVine
Abstract:
Estimating machine learning performance 'in the wild' is both an important and unsolved problem. In this paper, we seek to examine, understand, and predict the pointwise competence of classification models. Our contributions are twofold: First, we establish a statistically rigorous definition of competence that generalizes the common notion of classifier confidence; second, we present the ALICE (A…
▽ More
Estimating machine learning performance 'in the wild' is both an important and unsolved problem. In this paper, we seek to examine, understand, and predict the pointwise competence of classification models. Our contributions are twofold: First, we establish a statistically rigorous definition of competence that generalizes the common notion of classifier confidence; second, we present the ALICE (Accurate Layerwise Interpretable Competence Estimation) Score, a pointwise competence estimator for any classifier. By considering distributional, data, and model uncertainty, ALICE empirically shows accurate competence estimation in common failure situations such as class-imbalanced datasets, out-of-distribution datasets, and poorly trained models. Our contributions allow us to accurately predict the competence of any classification model given any input and error function. We compare our score with state-of-the-art confidence estimators such as model confidence and Trust Score, and show significant improvements in competence prediction over these methods on datasets such as DIGITS, CIFAR10, and CIFAR100.
△ Less
Submitted 24 October, 2019;
originally announced October 2019.
