16,489 Hits in 2.9 sec

Soft Calibration Objectives for Neural Networks [article]

Archit Karandikar, Nicholas Cain, Dustin Tran, Balaji Lakshminarayanan, Jonathon Shlens, Michael C. Mozer, Becca Roelofs
2021 arXiv   pre-print
However, deep neural networks are often under- or over-confident in their predictions.  ...  When incorporated post-training, the soft-binning-based calibration error objective improves upon temperature scaling, a popular recalibration method.  ...  Acknowledgements The authors thank Brennan McConnell and Mohammad Khajah who conducted initial explorations of soft binning calibration loss.  ... 
arXiv:2108.00106v2 fatcat:oxgu3qitdzb55b7lehp5ces5ii

A vision-based path planning and object tracking framework for 6-DOF robotic manipulator

Adil Shahzad, Xueshan Gao, Awais Yasin, Kamran Javed, Syed Muhammad Anwar
2020 IEEE Access  
ENSEMBLER The proposed framework coupled with the neural network increased robot accuracy for completion of the pick and place task for a soft object in real time.  ...  The neural network was a feed-forward neural network and the training was performed using stochastic gradient descent.  ... 
doi:10.1109/access.2020.3037540 fatcat:x3fnenthonhb3jetsgpmaufdgq

Large-Area and Low-Cost Force/Tactile Capacitive Sensor for Soft Robotic Applications

Amir Pagoli, Frédéric Chapelle, Juan-Antonio Corrales-Ramon, Youcef Mezouar, Yuri Lapusta
2022 Sensors  
A neural network architecture is used to calibrate the applied force with acceptable accuracy in the presence of noise, variation in gains, and non-linearity.  ...  This new sensor is very flexible and easy to fabricate, making it an appropriate choice for soft robot applications.  ...  Artificial neural network flowchart for calibrating the proposed sensor. Figure 10 . 10 Figure 10. The proposed two-layer feedforward network to calibrate the soft sensor.  ... 
doi:10.3390/s22114083 pmid:35684706 pmcid:PMC9185300 fatcat:3hgg7cb5o5d57e3kj6egd7i2aa

Review of machine learning methods in soft robotics

Daekyum Kim, Sang-Hun Kim, Taekyoung Kim, Brian Byunghyun Kang, Minhyuk Lee, Wookeun Park, Subyeong Ku, DongWook Kim, Junghan Kwon, Hochang Lee, Joonbum Bae, Yong-Lae Park (+2 others)
2021 PLoS ONE  
followed by a summary of the existing machine learning methods for soft robots.  ...  However, compared to rigid robots, soft robots have issues in modeling, calibration, and control in that the innate characteristics of the soft materials can cause complex behaviors due to non-linearity  ...  For the calibration of current soft sensors, a recurrent neural network (RNN), a deep learning algorithm specialized in time-series data, are frequently employed.  ... 
doi:10.1371/journal.pone.0246102 pmid:33600496 pmcid:PMC7891779 fatcat:alu4zm72irespj6wydikzjb6ie

Estimation of Tool-Tissue Forces in Robot-Assisted Minimally Invasive Surgery Using Neural Networks

Sajeeva Abeywardena, Qiaodi Yuan, Antonia Tzemanaki, Efi Psomopoulou, Leonidas Droukas, Chris Melhuish, Sanja Dogramadzi
2019 Frontiers in Robotics and AI  
The proposed method utilizes the current of the motors of the surgical instrument and neural network methods to estimate the force interaction.  ...  Validation for Using a Neural Network For a neural network methodology to be feasible, a relationship between the inputs and outputs of the network is required.  ...  TABLE 2 | 2 Maximum current for each grasp of hard and soft object, respectively.  ... 
doi:10.3389/frobt.2019.00056 pmid:33501071 pmcid:PMC7805956 fatcat:a42rbylx6fgrzl2yz55xc7dgg4

Synergy of modeling processes in the area of soft and hard modeling

Robert Sika, Jakub Hajkowski, I. Bondrea, C. Simion, M. Inţă
2017 MATEC Web of Conferences  
Collected data was used as an input to prepare soft model with the use of MLP (Multi-Layer Perceptron) neural network regression model.  ...  High complexity of production processes results in more frequent use of computer systems for their modeling and simulation.  ...  The aim is to prepare a soft mathematical model in form of a neural network, which can be asked in a much shorter time than using the Calibrate module (instead 120 seconds it is less than 5 seconds).  ... 
doi:10.1051/matecconf/201712104009 fatcat:y3rqusgf6fhxdn4ojnhei5irfu

Learn to grasp utilizing anthropomorphic fingertips together with a vision sensor

Y. Tada, K. Hosoda, M. Asada
2005 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems  
The sensor for the fingertip is difficult to calibrate because the sensor receptors are embedded randomly in the soft material.  ...  A robot should have softness and many sensors to manipulate an object dexterously and to adapt various environments.  ...  ACKNOWLEDGMENT This study was performed through the Advanced and Innovational Research Program in Life Sciences, and Grantin-Aid for Scientific Research (B) from the Ministry of Education, Culture, Sports  ... 
doi:10.1109/iros.2005.1545028 dblp:conf/iros/TadaHA05 fatcat:watv63kk7fh7letrz36yiit67u

Exoskeleton-covered soft finger with vision-based proprioception and tactile sensing [article]

Yu She, Sandra Q. Liu, Peiyu Yu, Edward Adelson
2020 arXiv   pre-print
We train neural networks for proprioception and shape (box versus cylinder) classification using data from the embedded sensors.  ...  These proposed techniques offer soft robots the high-level ability to simultaneously perceive their proprioceptive state and peripheral environment, providing potential solutions for soft robots to solve  ...  The authors would also like to thank Branden Robert Romero and Shaoxiong Wang for their help in hardware manufacturing and algorithm development, and Achu Wilson for his help in setting up the two Raspberry  ... 
arXiv:1910.01287v2 fatcat:nr2judpmjzdk3pbm5emwsehmry

Applications of Soft Computing in Civil Engineering: A Review

Vinay Chandwani, Vinay Agrawal, Ravindra Nagar
2013 International Journal of Computer Applications  
Soft Computing being a multi-disciplinary field uses a variety of statistical, probabilistic and optimization tools which complement each other to produce its three main branches viz., Neural Networks,  ...  The review paper presents the applications of two major Soft Computing techniques viz., Artificial Neural Networks and Genetic Algorithms in the field of Civil Engineering, which to some extent has replaced  ...  Among the forerunners in the field of soft computing is the Artificial Neural Network (ANN).  ... 
doi:10.5120/14047-2210 fatcat:ed46xbfhufbdjpdaal6zk4ozoe

Field programmable gate array implementation of an intelligent soft calibration technique for linear variable differential transformers

Santhosh Krishnan Venkata, Binoy Krishna Roy, Preeti Mohanty
2022 F1000Research  
The presented calibration technique using optimized neural network algorithms has produced reliable measurements as desired.  ...  The optimized neural network algorithm is implemented on a Field Programmable Gate Array (FPGA) chip for testing and validation in real life.  ...  This paper describes an intelligent adaptive calibration technique using an optimized Artificial Neural Network (ANN) for displacement measurements by LVDT.  ... 
doi:10.12688/f1000research.110093.1 fatcat:lmvctps7cngm7f3hgfaf3mimvq

Soft Computing Approach for Rainfall-runoff Modelling: A Review

Vinay Chandwani, Sunil Kumar Vyas, Vinay Agrawal, Gunwant Sharma
2015 Aquatic procedia  
Soft Computing being a multi-disciplinary field uses a variety of statistical, probabilistic and optimization tools which complement each other to produce its three main branches viz., Neural Networks,  ...  Enormous cost and manpower utilization encountered in constructing a water resource project demands a great deal of attention in devising precise Rainfall-Runoff models for its successful performance.  ...  Jeong and Kim (2005) used two neural network models viz., single neural network (SNN) and ensemble neural network (ENN) for providing better rainfall-runoff simulation.  ... 
doi:10.1016/j.aqpro.2015.02.133 fatcat:5gksb6guqrer5fkdbtysgjkffm

A Neuro-Fuzzy Model with SEER-SEM for Software Effort Estimation [article]

Wei Lin Du, Danny Ho, Luiz Fernando Capretz
2015 arXiv   pre-print
In order to address this issue, a novel soft computing framework was previously developed.  ...  Furthermore, the results of this research indicate that the general neuro-fuzzy framework can work with various algorithmic models for improving the performance of software effort estimation.  ...  The objectives of this model are to improve the FP complexity weight systems by fuzzy logic, to calibrate the weight values of the unadjusted FP through the neural network, and to produce a calibrated  ... 
arXiv:1508.00032v1 fatcat:agv6ncp5hbbxfgt2zkjuisinga

Calibrated Adversarial Training [article]

Tianjin Huang, Vlado Menkovski, Yulong Pei, Mykola Pechenizkiy
2021 arXiv   pre-print
We provide theoretical analysis on the calibrated robust error and derive an upper bound for it.  ...  The method produces pixel-level adaptations to the perturbations based on novel calibrated robust error.  ...  The architectures of auxiliary neural network g ϕ for MNIST and CIFAR-can be found in Appendix C.  ... 
arXiv:2110.00623v2 fatcat:tvqzw5tgzffrle4l4etdjezxle

Dropout Distillation for Efficiently Estimating Model Confidence [article]

Corina Gurau, Alex Bewley, Ingmar Posner
2018 arXiv   pre-print
We propose an efficient way to output better calibrated uncertainty scores from neural networks.  ...  We also propose better calibration within the state of the art Faster R-CNN object detection framework and show, using the COCO dataset, that DDN helps train better calibrated object detectors.  ...  For a non-linear neural network however, the term p(w|D) is analytically intractable.  ... 
arXiv:1809.10562v1 fatcat:x6mr7mqgovfjtpj6a5sly4j6ca

An intuitive view to compare intelligent systems

A.R. Nazemi, M.R.T. Akbarzadeh, S.M. Hosseini
2004 IEEE Annual Meeting of the Fuzzy Information, 2004. Processing NAFIPS '04.  
The results shows that the modeling efficiency increases by moving toward neural modeling; particularly, the fuzzy clustering-based neural network is the most efficient and accurate paradigm among performed  ...  ., Rainfall-Runoff modeling is introduced and nine soft computing-based modeling approaches are considered to describe the rainfall-runoff process in a particular case study.  ...  Hsu for providing data of Leaf River basin and some of their publications.  ... 
doi:10.1109/nafips.2004.1337363 fatcat:tegj5466ejgank45wkoetjush4
« Previous Showing results 1 — 15 out of 16,489 results