Artificial Intelligence (AI) & Cognitive Sensing at the Endpoint
Introducing a Complete AI Platform Solution
Supported by an Ecosystem of partners
Enables AI at the Endpoint device
Includes Sensor Processing platform, Neurons for AI computing, and Application SW for Knowledge building of data and classification
Scaling across bigger AI systems
No need for in-house expertise of data analytics, DSP Processing, App Coding
Predictive Maintenance applications
Vision Inspection System
Endpoint Maker Opportunity
365M industrial devices to rely on cognitive computing in 2023
Video applications to remain predominant, with strong growth in manufacturing and retail
Only 25% of devices to rely on the cloud for training, inference, or both in 2023
Addressing the Challenges of Endpoint
Organizations working at the endpoint don’t know to how make use of big data, cloud or AI
Lack of data scientist and data modeling know-how
Diversity of different applications and use cases makes defining model and algorithms for every use case difficult to buy off the shelf
Development cost and complexity of managing deployed models on distributed devices
Bringing together key ecosystem partners with various expertise to enable OEMs to develop AI solutions for the Endpoint
QuickAI™ Platform Solution
The new QuickAI platform provides an all-inclusive low power solution and development environment to economically incorporate the benefits of AI in endpoint applications. It features technology, software, and toolkits from General Vision, Nepes, SensiML and QuickLogic, all of which have formed a tightly-coupled ecosystem to solve the challenges associated with the implementation of AI for endpoint applications.
Application SW to make AI scalable
Collecting and training of data
Conducts both the learning and the inference
Brings Scaling across applications and customers
NeuroMem IP – Radial Basis Function (RBF)
Concurrent, parallel with scalability (cascade design for additional neurons)
General Vision: The QuickAI platform is based on General Vision's NeuroMem neural network IP, which has been licensed by Nepes and integrated into the Nepes Neuromorphic NM500 AI learning device. Both General Vision and Nepes provide software for configuring and training the neurons in the network. In addition, SensiML provides an analytics toolkit to quickly and easily build smart sensor algorithms for endpoint IoT applications.
General Vision's technology is the foundation element of the platform. This technology enables on-chip exact and fuzzy pattern matching and learning using a scalable parallel architecture of Radial Basis Function neurons. The parallel architecture results in fixed latency for any number of neurons and a very low, energy efficient operating frequency. General Vision supplies the Knowledge Builder tool suite and SDK used to train and configure the neurons in a NeuroMem network.
Nepes: The Nepes Neuromorphic NM500 implements the NeuroMem technology in an energy efficient, small form factor component. This AI-enabling component can be trained in the field to recognize patterns in real time, and multiple devices can be chained to provide any number of neurons. In addition to the NM500, Nepes supplies the Knowledge Studio software tools used for configuring and training the neurons in the NM500 device.
SensiML: SensiML complements the General Vision/Nepes technology by providing the SensiML Analytics Toolkit, which simplifies the task of generating endpoint AI solutions by providing tools that automate the management of training data, optimize the choice of feature extraction algorithms, and automate code generation for the resulting AI solution.
QuickLogic: The EOS™ S3 voice and sensor processing platform from QuickLogic is the final critical element of this new platform initiative. Its ultra-low power, sophisticated audio and sensor processing and embedded FPGA make it the perfect host for the NM500 and the software that implements AI solutions using the NM500.
The platform includes the QuickAI Hardware Development Kit (HDK) with EOS S3, two NM500 devices, accelerometer, gyroscope, magnetometer, microphones, Nordic Bluetooth® Low Energy device, flash memory, and an Intel Edison-compatible connector that allows access to Edison daughter boards such as uSD.
QuickAI HDK Platform
Benefits of QuickAI HDK:
Demo/Eval/Development /Deployment Platform
Time to Development & Time to Market
Enables AI (Data Collection, Feature Extraction, Classifier) with Motion, Acoustic & Image Processing