Market and Technology Trends

Neuromorphic Computing and Sensing 2021

By Yole Intelligence — May 2021

Industrial, consumer, and automotive applications are driving the adoption of neuromorphic computing and sensing technologies. The first products are now hitting the market.

What's new :

New applications are included in the market forecast, such as consumer security cameras, smartwatches, earbuds, consumer AR/VR, machine vision for industrial robotics, industrial
Extended analysis for computing, software, and memory technologies
Description of use cases for neuromorphic technologies
Deeper and broader analysis of the ecosystem and related products

THERE IS A STRONG NEED FOR NEUROMORPHIC TECHNOLOGIES

Artificial Intelligence (AI) is hungry for performance, and Moore’s law dynamics will not suffice to cover the needs of the ongoing 5G/IoT/AR/Robotic revolution. Brute force is currently being used to leverage the power of AI, but this approach is not scalable. It will hit a heat wall, a data wall, and a cost wall related to the ability of the semiconductor industry to deliver at a certain pace (Moore’s law) the incremental cost to performance improvements. Current AI computing will not deliver, so what AI technology would be necessary? Neuromorphic computing and sensing solutions, drawing inspiration from what happens in the brain, have key specificities to compete within the existing AI landscape and constraints. These technologies will address most of the current challenges and could represent 20% of all AI computing & sensing by 2035.
Industrial applications will be the first to use neuromorphic technologies, driven by high speed, low latency, and offline learning enabling more autonomous features and performance. Players such as Prophesee, Brainchip, and Nepes AI/General Vision already have products in the market targeting industrial applications, and more players will follow in the coming years. The consumer market will also benefit from neuromorphic technologies enabling more AI applications at the edge on battery-powered devices, ensuring privacy and safety of personal data.
In the automotive market, increased integration of neuromorphic technologies will leverage the extremely low latency and power efficiency. While it will take longer for neuromorphics to be adopted in this promising market due to the long product cycle, some projects are already being announced, such as Xperi’s Driver Monitoring System and Terranet’s ADAS cameras and laser.
Additionally, the cloud server market could also benefit from neuromorphic computing technologies, leveraging low latency and online learning to improve the performance of applications such as cybersecurity and fraud detection. The considerable power efficiency could also help to limit the power consumption growth in data centers which is a growing concern. Large players such as Intel and IBM are already creating neuromorphic server prototypes by assembling their massively scalable Loihi and TrueNorth chips, respectively.

Current neuromorphic device architectures can also vary significantly with respect to the organization of the memory and computing components on the silicon chips. There is currently a clear trend towards “in-memory-computing” solutions: several companies are developing designs with mainstream embedded memory (e.g., SRAM) distributed across cores or neurons; various players are also considering the adoption of emerging non-volatile memory (NVM) elements assembled in crossbar arrays, leveraging the “synapsis-like” properties of resistive memories (e.g., PCM, OxRAM, CBRAM). Such memristive technologies are currently in development and could become viable solutions for edge neuromorphic devices by 2030.

A SOLID AND DIVERSIFIED ECOSYSTEM IS GROWING

The neuromorphic ecosystem is today very dynamic with three main categories of players: university & research institute, labs affiliated with large companies, and start-ups. The latter are the first players to bring products to the market for edge computing, targeting industrial, automotive, and consumer applications. They will test new approaches in the market in a real-life environment.
From their side, universities are forming extensive collaborations, often supported by governments, to develop the technology and understand the hard-science potential. This covers a vast range, starting from a simulation of the brain on a silicon chip, and involves partnerships with various companies to develop proofs-of-concept in the field. Labs affiliated with large companies are deeply involved in these collaborations and often take the lead. Intel and IBM developed neuromorphic communities around their chips to help the software ecosystem grow, increase the maturity of neuromorphic AI, and test use cases directly with application players.

OVER THE PAST YEARS, SEVERAL KEY EVENTS HAVE OCCURRED

Many start-ups raised funds, such as Brainchip’s $21.5M in 2017, Prophesee’s $28M in 2019, and GrAI Matter Labs’ $29M in 2018 and 2020 combined. This was followed by the acquisition of several start-ups, such as Insightness by Sony and Cele Pixel by Hua Capital (OmniVision) in 2020.
Many products were also announced, such as Intel’s Loihi in 2017 and Brainchip’s Akida in 2018. Sony and Prophesee released an HD neuromorphic sensor in 2019, and Intel released Pohoiki Springs in 2020, reaching 100M neurons, which is currently available for researchers with a simple connection. Some companies will soon start hitting the market; Brainchip begins volume production of the Akida AI processor in April 2021, and the demonstration kit of Prophesee & Sony’s HD sensor will be made available in May 2021.
Finally, many projects with application players have also been announced. For example, Terranet and Daimler jointly released a neuromorphic ADAS solution in February 2021, and Xperi and Prophesee released an in-cabin monitoring solution in April 2021.
With regard to this accelerating surge of energy, we forecast that neuromorphic technologies will achieve enough maturity for a market take-off by 2025, first in the industrial and consumer market segments then quickly followed by Automotive.

Report's key features:

Definition and description of neuromorphic technologies
2020 - 2035 Neuromorphic sensing and computing market forecast in $
2020 - 2035 Neuromorphic technology penetration in %
Mobile, automotive, medical, industrial and data center market trends
Neuromorphic computing and in-memory computing players – mapping and technology segmentation
Neuromorphic image sensor player mapping
Image sensor technology trends
Available neuromorphic processors and image sensors
Neuromorphic software description, software readiness, and acceptance
Neuromorphic player's descriptions
Neuromorphic in-memory computing concept
Emerging non-volatile memory, ecosystem, and roadmap
Forecast for emerging non-volatile memory
Embedded non-volatile memory for in-memory computing

Objectives of the report:

To provide market data on key neuromorphic adoption areas with details on:
- semiconductor-level revenue forecasts, shipments, ASPs, and penetration rate of neuromorphic sensing and computing for 13 different applications
- market dynamics and segmentation breakdown, split by application, market, and end-system
- the two axes of focus: sensing and computing
- memory architecture and related technology trends
To deliver an in-depth understanding of the neuromorphic ecosystem:
- Who are the key CIS manufacturers, AI accelerator manufacturers, and eNVM manufacturers, and how are they related?
- Who are the key suppliers to watch?
- How will the technology and associated market evolve?
To present key technical insights and analyses regarding future technology trends and challenges:
- Manufacturing technologies and structural design
- Device technologies and applications across the different markets
- Technology insights for game-changing applications like deep learning and neuromorphic approaches

To provide market data on key neuromorphic adoption areas with details on:

semiconductor-level revenue forecasts, shipments, ASPs, and penetration rate of neuromorphic sensing and computing for 13 different applications
market dynamics and segmentation breakdown, split by application, market, and end-system
the two axes of focus: sensing and computing
memory architecture and related technology trends

semiconductor-level revenue forecasts, shipments, ASPs, and penetration rate of neuromorphic sensing and computing for 13 different applications
market dynamics and segmentation breakdown, split by application, market, and end-system
the two axes of focus: sensing and computing
memory architecture and related technology trends

To deliver an in-depth understanding of the neuromorphic ecosystem:

Who are the key CIS manufacturers, AI accelerator manufacturers, and eNVM manufacturers, and how are they related?
Who are the key suppliers to watch?
How will the technology and associated market evolve?

Who are the key CIS manufacturers, AI accelerator manufacturers, and eNVM manufacturers, and how are they related?
Who are the key suppliers to watch?
How will the technology and associated market evolve?

To present key technical insights and analyses regarding future technology trends and challenges:

Manufacturing technologies and structural design
Device technologies and applications across the different markets
Technology insights for game-changing applications like deep learning and neuromorphic approaches

Manufacturing technologies and structural design
Device technologies and applications across the different markets
Technology insights for game-changing applications like deep learning and neuromorphic approaches

Do you have an account?

Search