2,207 Hits in 4.2 sec

Fault detection in autonomous robots based on fault injection and learning

Anders Lyhne Christensen, Rehan O'Grady, Mauro Birattari, Marco Dorigo
2007 Autonomous Robots  
In this paper, we study a new approach to fault detection for autonomous robots.  ...  We use back-propagation neural networks to synthesize fault detection components based on the data collected in the training runs.  ...  The three experimental setups We have chosen three setups in which to study fault detection based on fault injection and learning.  ... 
doi:10.1007/s10514-007-9060-9 fatcat:rxib5jy6xrcznd3zutmx3g34me

Automatic Synthesis of Fault Detection Modules for Mobile Robots

A.L. Christensen, R. O'Grady, M. Birattari, M. Dorigo
2007 Second NASA/ESA Conference on Adaptive Hardware and Systems (AHS 2007)  
In this paper, we present a new approach for automatic synthesis of fault detection modules for autonomous mobile robots.  ...  In each experiment, we record all sensory inputs from the robots while they are operating normally and after software-simulated faults have been injected.  ...  Fault detection for autonomous robots involves classification based on partial and imperfect information due to limited and noisy sensors and actuators.  ... 
doi:10.1109/ahs.2007.37 dblp:conf/ahs/ChristensenOBD07 fatcat:mdhv2ut4wraelmnansodpdopui

Manipulation of Camera Sensor Data via Fault Injection for Anomaly Detection Studies in Verification and Validation Activities For AI [article]

Alim Kerem Erdogmus, Mustafa Karaca, Assist. Prof. Dr. Ugur Yayan
2022 arXiv   pre-print
The study is based on an existing camera fault injection software that injects faults into the cameras of a working robot and collects the normal and faulty images recorded during this injection.  ...  The database obtained in the study is a source for the detection of anomalies that may occur in robotic systems.  ...  The created database is designed to be a source for artificial intelligence studies that will enable the detection of image faults that may occur in robot cameras. 2 Creation of Camera Fault Injection  ... 
arXiv:2108.13803v4 fatcat:gv7taqteubcxdow5xa2fb2cgei

Fault Tolerant Planning for Critical Robots

Benjamin Lussier, Matthieu Gallien, Jeremie Guiochet, Felix Ingrand, Marc-Olivier Killijian, David Powell
2007 37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN'07)  
The paper presents an implementation of these mechanisms on an existing autonomous robot architecture, and evaluates their impact on performance and reliability through the use of fault injection.  ...  The proposed mechanisms focus on development faults in planning models and heuristics, through the use of diversification.  ...  Finally, we validate the proposed mechanisms through an experimental framework based on fault injection.  ... 
doi:10.1109/dsn.2007.50 dblp:conf/dsn/LussierGGIKP07 fatcat:uriy4oqfcjbbrftb3sjtmu3dem

The Challenges and Opportunities of Human-Centered AI for Trustworthy Robots and Autonomous Systems [article]

Hongmei He, John Gray, Angelo Cangelosi, Qinggang Meng, T.Martin McGinnity, Jörn Mehnen
2021 arXiv   pre-print
The trustworthiness of Robots and Autonomous Systems (RAS) has gained a prominent position on many research agendas towards fully autonomous systems.  ...  RAS must be (i) safe in any uncertain and dynamic surrounding environments; (ii) secure, thus protecting itself from any cyber-threats; (iii) healthy with fault tolerance; (iv) trusted and easy to use  ...  [71] categorised three types of fault diagnosis and prognosis in predictive maintenance systems, such as knowledge based, traditional ML based and Deep Learning based approaches.  ... 
arXiv:2105.04408v1 fatcat:jvlx7lkjizgnbcu2t27ndi4l3q

Analyzing and Improving Fault Tolerance of Learning-Based Navigation Systems [article]

Zishen Wan, Aqeel Anwar, Yu-Shun Hsiao, Tianyu Jia, Vijay Janapa Reddi, Arijit Raychowdhury
2021 arXiv   pre-print
Learning-based navigation systems are widely used in autonomous applications, such as robotics, unmanned vehicles and drones.  ...  We further propose two efficient fault mitigation techniques that achieve 2x success rate and 39% quality-of-flight improvement in learning-based navigation systems.  ...  This work was supported in part by C-BRIC and ADA, two of six centers in JUMP, a Semiconductor Research Corporation (SRC) program sponsored by DARPA.  ... 
arXiv:2111.04957v1 fatcat:kxb7rbihkjfr7mdkutz2q25nre

The impact of the soft errors in convolutional neural network on GPUs: Alexnet as case study

Khalid Adam, Izzeldin I. Mohd, Younis M. Younis
2021 Procedia Computer Science  
Convolutional Neural Networks (CNNs) have been increasingly deployed in many applications, including safety critical system such as healthcare and autonomous vehicles.  ...  Results show that FADD and LD are the top vulnerable instructions against soft errors for Alexnet model, both instructions generate at least 84% of injected faults as SDC errors.  ...  It is worth mentioning that CC and PR instructions are one-bit operations. In other words, they only perform toggle between zero and one, thus, we can only inject single-bit-flip faults into them.  ... 
doi:10.1016/j.procs.2021.02.012 fatcat:jwloam4rcbhbpncybbdmwwszzi

Surgeon Training in Telerobotic Surgery via a Hardware-in-the-Loop Simulator

Xiao Li, Homa Alemzadeh, Daniel Chen, Zbigniew Kalbarczyk, Ravishankar K. Iyer, Thenkurussi Kesavadas
2017 Journal of Healthcare Engineering  
The proposed simulator is built upon the Raven-II™ open source surgical robot, integrated with a physics engine and a safety hazard injection engine.  ...  Also, a Fast Marching Tree-based motion planning algorithm is used to help trainee learn the optimal instrument motion patterns.  ...  CNS 13-14891 and CNS 15-45069 and a grant through the JUMP-ARCHES (Applied Research for Community Health through Engineering and Simulation) program for addressing safety and reliability of surgical robots  ... 
doi:10.1155/2017/6702919 pmid:29065635 pmcid:PMC5560083 fatcat:itypkorevngkjjs2w7khsbaehu

Online and Offline Diagnosis of Motor Power Cables Based on 1D CNN and Periodic Burst Signal Injection

Heonkook Kim, Hyeyun Jeong, Hojin Lee, Sang Woo Kim
2021 Sensors  
We introduce a new approach for online and offline soft fault diagnosis in motor power cables, utilizing periodic burst injection and nonintrusive capacitive coupling.  ...  Both online and offline diagnoses with on-site diagnostic ability are needed because the equipment in the automated lines operates for 24 h per day, except during scheduled maintenance.  ...  Data Availability Statement: The data presented in this study are available on request from the corresponding author.  ... 
doi:10.3390/s21175936 pmid:34502827 fatcat:mdyufiqhlbbgtljt4afdh73qt4

Adaptive Online Fault Diagnosis in Autonomous Robot Swarms

James O'Keeffe, Danesh Tarapore, Alan G. Millard, Jon Timmis
2018 Frontiers in Robotics and AI  
A case has been made for an active approach to fault tolerance in swarm robotic systems, whereby the swarm can identify and resolve faults that occur during operation.  ...  The results of our simulated experiments show that our system is flexible, scalable, and improves swarm tolerance to various electro-mechanical faults in the cases examined.  ...  AUTHOR CONTRIBUTIONS JO: performed all experiments and obtained all results described; Principal author of this paper; DT: critical revisions; AM: critical revisions; JT: critical revisions.  ... 
doi:10.3389/frobt.2018.00131 pmid:33501009 pmcid:PMC7805982 fatcat:ostfo74hpje77ak4jdaqmgyakm

Towards Stochastic Fault-tolerant Control using Precision Learning and Active Inference [article]

Mohamed Baioumy, Corrado Pezzato, Carlos Hernandez Corbato, Nick Hawes, Riccardo Ferrari
2021 arXiv   pre-print
This work presents a fault-tolerant control scheme for sensory faults in robotic manipulators based on active inference.  ...  We propose a stochastic fault-tolerant scheme based on active inference and precision learning which does not require a priori threshold definitions to trigger fault recovery.  ...  Finally, precision learning performs stochastic fault-detection rather than deterministic. Most importantly, this approach based on precision learning can be improved in many ways.  ... 
arXiv:2109.05870v1 fatcat:q5gj37gcpje7nlhjhvzdmdegna

Evolutionary online behaviour learning and adaptation in real robots

Fernando Silva, Luís Correia, Anders Lyhne Christensen
2017 Royal Society Open Science  
Online evolution of behavioural control on real robots is an open-ended approach to autonomous learning and adaptation: robots have the potential to automatically learn new tasks and to adapt to changes  ...  In this article, we successfully evolve neural network-based controllers in real robotic hardware to solve two single-robot tasks and one collective robotics task.  ...  FS, 0000-0003-2702-9295 Online evolution of behavioural control on real robots is an open-ended approach to autonomous learning and adaptation: robots have the potential to automatically learn new tasks  ... 
doi:10.1098/rsos.160938 pmid:28791130 pmcid:PMC5541525 fatcat:gffdzsmbznecbmzxnq4a3wsi6q

Beta Residuals: Improving Fault-Tolerant Control for Sensory Faults via Bayesian Inference and Precision Learning [article]

Mohamed Baioumy, William Hartemink, Riccardo M.G. Ferrari, Nick Hawes
2022 arXiv   pre-print
In this paper, we introduce a precision-learning based Bayesian FTC approach and a novel beta residual for fault detection.  ...  Model-based fault-tolerant control (FTC) often consists of two distinct steps: fault detection & isolation (FDI), and fault accommodation.  ...  In the second case, we use an existing FDI method based on the probabilisticallyrobust thresholds, then when a fault is detected, precision learning is triggered.  ... 
arXiv:2204.08035v1 fatcat:fiwhb37t35codmqbfawondflpe

Run-time detection of faults in autonomous mobile robots based on the comparison of simulated and real robot behaviour

Alan G. Millard, Jon Timmis, Alan F.T. Winfield
2014 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems  
2014) Run-time detection of faults in autonomous mobile robots based on the comparison of simulated and real robot behaviour.  ...  In [4] we proposed a novel method of exogenous fault detection capable of detecting partial failures, based on the comparison of expected and observed robot behaviour.  ...  CONCLUSIONS & FUTURE WORK To our knowledge, this paper represents the first application of simulation to predicting robot behaviour for the detection of faults in autonomous mobile robots.  ... 
doi:10.1109/iros.2014.6943084 dblp:conf/iros/MillardTW14 fatcat:bxdgryqa5jhyje5rl2gwtqpyou

Real-Time Context-aware Detection of Unsafe Events in Robot-Assisted Surgery [article]

Mohammad Samin Yasar, Homa Alemzadeh
2020 arXiv   pre-print
However, with increasing complexity and connectivity of software and major involvement of human operators in the supervision of surgical robots, there remain significant challenges in ensuring patient  ...  Our experiments using data from two surgical platforms show that the proposed system can detect unsafe events caused by accidental or malicious faults within an average reaction time window of 1,693 milliseconds  ...  and a fault injection tool that mimics the effect of technical faults and attacks in the robot control system.  ... 
arXiv:2005.03611v2 fatcat:i5h63kf6wjcfnisa337x2ub56a
« Previous Showing results 1 — 15 out of 2,207 results