Learning sensorimotor control with neuromorphic sensors: Toward hyperdimensional active perception.
Anton Mitrokhin,Peter Sutor,Cornelia Fermüller,Yiannis Aloimonos +3 more
- Vol. 4, Iss: 30
TLDR
This work used DVS for visual perception and showed that the visual component can be bound with the system velocity to enable dynamic world perception, which creates an opportunity for real-time navigation and obstacle avoidance.Abstract:
The hallmark of modern robotics is the ability to directly fuse the platform's perception with its motoric ability-the concept often referred to as "active perception." Nevertheless, we find that action and perception are often kept in separated spaces, which is a consequence of traditional vision being frame based and only existing in the moment and motion being a continuous entity. This bridge is crossed by the dynamic vision sensor (DVS), a neuromorphic camera that can see the motion. We propose a method of encoding actions and perceptions together into a single space that is meaningful, semantically informed, and consistent by using hyperdimensional binary vectors (HBVs). We used DVS for visual perception and showed that the visual component can be bound with the system velocity to enable dynamic world perception, which creates an opportunity for real-time navigation and obstacle avoidance. Actions performed by an agent are directly bound to the perceptions experienced to form its own "memory." Furthermore, because HBVs can encode entire histories of actions and perceptions-from atomic to arbitrary sequences-as constant-sized vectors, autoassociative memory was combined with deep learning paradigms for controls. We demonstrate these properties on a quadcopter drone ego-motion inference task and the MVSEC (multivehicle stereo event camera) dataset.read more
Citations
More filters
Journal ArticleDOI
In-memory hyperdimensional computing
Geethan Karunaratne,Geethan Karunaratne,Manuel Le Gallo,Giovanni Cherubini,Luca Benini,Abbas Rahimi,Abu Sebastian +6 more
TL;DR: In this article, a complete in-memory hyperdimensional computing system is presented, where all operations are implemented on two memristive crossbar engines together with peripheral digital complementary metaloxide-semiconductor (CMOS) circuits.
Journal ArticleDOI
Classification Using Hyperdimensional Computing: A Review
Lulu Ge,Keshab K. Parhi +1 more
TL;DR: Hyperdimensional (HD) computing as discussed by the authors is built upon its unique data type referred to as hypervectors, which is typically in the range of tens of thousands of dimensions and is used to solve cognitive tasks.
Journal ArticleDOI
Plasmonic Alloys Reveal a Distinct Metabolic Phenotype of Early Gastric Cancer.
Haiyang Su,Xinxing Li,Lin Huang,Jing Cao,Mengji Zhang,Vadanasundari Vedarethinam,Wen Di,Zhiqian Hu,Kun Qian +8 more
TL;DR: In this paper, mesoporous PdPtAu alloys are designed to characterize the metabolic profiles in human blood, and a distinct metabolic phenotype is revealed for early GC by sparse learning, resulting in precise GC diagnosis with an area under the curve of 0.942.
Journal ArticleDOI
A Programmable Hyper-Dimensional Processor Architecture for Human-Centric IoT
TL;DR: A complete, programmable architecture for ultra energy-efficient supervised classification using HD computing, and the main architectural decisions for similar systems harnessing variability in emerging devices (eg. CNFET and RRAM) are established.
Proceedings ArticleDOI
OnlineHD: Robust, Efficient, and Single-Pass Online Learning Using Hyperdimensional System
TL;DR: OnlineHD as discussed by the authors proposes an adaptive hyper-dimensional computing (HDC) training framework for accurate, efficient, and robust learning, which identifies common patterns and eliminates model saturation during single-pass training.
References
More filters
Proceedings ArticleDOI
Glove: Global Vectors for Word Representation
TL;DR: A new global logbilinear regression model that combines the advantages of the two major model families in the literature: global matrix factorization and local context window methods and produces a vector space with meaningful substructure.
Posted Content
Efficient Estimation of Word Representations in Vector Space
TL;DR: This paper proposed two novel model architectures for computing continuous vector representations of words from very large data sets, and the quality of these representations is measured in a word similarity task and the results are compared to the previously best performing techniques based on different types of neural networks.
Posted Content
End to End Learning for Self-Driving Cars
Mariusz Bojarski,Davide Del Testa,Daniel Dworakowski,Bernhard Firner,Beat Flepp,Prasoon Goyal,Lawrence D. Jackel,Mathew Monfort,Urs A. Muller,Jiakai Zhang,Xin Zhang,Jake Zhao,Karol Zieba +12 more
TL;DR: A convolutional neural network is trained to map raw pixels from a single front-facing camera directly to steering commands and it is argued that this will eventually lead to better performance and smaller systems.
Book
Active vision
Andrew Blake,Alan L. Yuille +1 more
TL;DR: Sometimes, reading is very boring and it will take long time starting from getting the book and start reading, but in modern era, you can take the developing technology by utilizing the internet and search for the book that is needed.
Active perception
TL;DR: In this article, a search of such sequences of steps that would minimize a loss function while still seeking the most information is formulated as a search for the sequence of steps to minimize the loss function.