General Synaptics

Safety_Critical_Machine_Learning

Data Efficient Learning - Robust Automatic TArget Recognition (DEL - RATAR™)

Our algorithms and technologies are aimed at bringing robust automatic target recognition to the most isolated application contexts (the concept of the cloud is totally non-existent) and to the most extreme environments where it is often difficult to find large quantities of data.

Operating at the wet edge is worth drawing out some distinctions for a couple of reasons. It’s chaotic. It’s uncertain. It is challenging for sensors to gather high-fidelity information underwater. It is a wild environment with limited communication. And so, for that reason, it’s the perfect environment to operate our machine learning algorithms. We operate with one-shot learning algorithms and limited amounts of data that is not the world we live in at the moment.

At present, in addition to lethal mines, there are also many manmade devices with detection, inspection, and strike capabilities. How to accurately recognize underwater manmade equipment is one of the current key research directions.

In the process of recognizing dangerous targets, it not only needs the target be accurately recognized but also needs to calculate the target’s status information, such as position, movement direction, and travel speed.

Convolutional Neural Networks +

L-Anti-Spoofing norm Engineered Recognizer

Synthetic Aperture Sonars (SAS) images of two different types of mine

Exclusive Pattern Recognition Triple Version Algorithm

Hierarchical Pattern Recognition Triple Version Algorithm

Convolutional Neural Network + Pattern Recognition Triple Version Algorithm – Image of a submarine identification

Our technology uses deep-learning for feature extraction and specific final classifiers. Feature extraction can be achieved via transfer learning with DESERT™ (DEep SElf Reflection Transfer™) technology trained with heterogeneous data. All final stage classifiers are enabled to learn in a single cycle. We can also use the Neuromem scalable neuromorphic chip for applications that require pre-classification stages for filtering and pre-processing purposes on board the sensors.

One Shot Learning SW algorithms

One Shot Learning Neuromorphic Chip Technology

for SMART SENSORS applications

WhoWeAre WhatWeDo Technology

Customers Partners News

General Synaptics

Aerospace_and_Defence_Machine_Learning_Company

VAT NUMBER:_IT02670700992

REA NUMBER: GE-503104

NATO_CAGE_CODE: AK845

Email:_luca.marchese@synaptics.org