With the world situation changing continuously in 2022, some technology trends have stalled while others accelerated. Supply chain challenges, labor shortages and economic uncertainty have prompted companies to reevaluate their budgets for new technologies.
For many organizations, AI (artificial intelligence) is seen as a potential solution to deliver improved efficiency, differentiation, automation, and cost reduction.
Up until now, AI computing has worked almost exclusively in the cloud. But increasingly diverse streams of data are being generated around the clock from sensors at the edge. These require real-time inference, which is leading to more AI implementations moving to edge computing.
For airports, stores, hospitals, etc., AI application solutions offer high efficiency, automation, and even cost reduction, which is why adopting Edge AI technology Advanced has been continuously promoted.
In 2023, we will likely continue to witness a year of many challenges, which is expected to accelerate even more trends in Edge AI technology that will be covered below. .
1. Focus on AI use cases with high ROI
Return on investment has always been an important factor for the acquisition and implementation of new technologies. But with companies looking for new ways to reduce costs and gain a competitive advantage, AI projects will become more common.
A few years ago, AI was often seen as experimental, but according to research from IBM, 35% of companies today say they are already using AI in their business, and an additional 42% say they are search and discover AI solutions. In particular, Edge AI use cases can help increase efficiency and reduce costs, making them attractive places to focus new investments.
For example, supermarkets and large stores are investing heavily in AI at vending machines to reduce losses due to theft and human error. With solutions that can detect errors with 98% accuracy, companies can quickly see a return on investment in just a few months.
An AI-based industrial inspection solution can also deliver immediate results, enhancing inspection efficiency across factory lines. Powered by aggregate data, AI can detect errors at a much higher rate than humans and resolve many types of errors that cannot be captured manually, resulting in more accurate identification. , less errors.
2. Growth in human-machine cooperation
Often considered a far-fetched use case of Edge AI, the use of intelligent machines and automated robots is on the rise. From automated distribution facilities to meet same-day delivery needs, to robots that monitor grocery stores to detect excesses and out of stock, to robotic arms that work alongside humans people on the production line, these intelligent machines are becoming more and more popular.
According to Gartner, the use of robots and intelligent machines is expected to increase significantly by the end of the decade. “By 2030, 80% of humans will interact with smart robots on a daily basis, as smart robot advances in intelligence, social interaction and enhanced human capabilities, increase from less than 10% today.” (Gartner, “Emerging Technologies: AI Roadmap for Smart Robots — Journey to a Super Intelligent Humanoid Robot”, G00761328, June 2022)
For this future to become a reality, one area of focus to focus on in 2023 is supporting human-machine collaboration. Automated processes benefit from the power and repeatable actions performed by robots, helping humans to perform specialized and skillful tasks better suited to our skills. It is hoped that organizations will invest more in this human-machine collaboration by 2023 as a way to alleviate labor shortages and supply chain problems.
3. New AI use cases with safety assurance
Related to the trend of human-machine cooperation is the trend of improving safety with AI. First seen in self-driving vehicles, many companies are looking to use AI to add proactive and flexible safety measures to industrial environments.
In the past, the safety function has been applied in the industrial environment in a binary way, with the main role of the safety function being to immediately prevent the device from causing any harm. any harm or damage when an event is triggered. AI, on the other hand, works in conjunction with context perception to predict an ongoing event. This allows AI to proactively send alerts about potential future safety-related events, preventing events before they happen, which can greatly reduce safety incidents and reduce downtime. related downtime in an industrial environment.
New safety functionality standards defining the use of AI in safety are expected to be released in 2023 and will open the door to early adoption in factories, warehouses, and agricultural use cases. , etc. One of the first areas of AI safety adoption will focus on improving worker safety, including worker posture detection, falling object prevention, and personal protective equipment detection. .
4. IT focuses on cybersecurity at the edge
Cyberattacks increased by 50% in 2021 and have not decreased since, making this a top focus for IT organizations. Edge computing, especially when combined with AI use cases, can increase cybersecurity risks for many organizations by creating a wider attack surface outside of the traditional data center and its firewall.
Edge AI in industries like manufacturing, energy, and transportation requires IT teams to extend their security reach to environments traditionally managed by operational technology teams. Operations technology teams often focus on operational efficiency as their primary metric, relying on live systems without a network connection to the outside world. Edge AI use cases will begin to circumvent these limitations, requiring IT to enable cloud connectivity while maintaining strict security standards.
With billions of devices and sensors around the world to be connected to the internet, IT organizations must both protect edge devices from direct attack and consider cloud and network security. In 2023, expect to see AI applied to cybersecurity. Log data generated from IoT networks can now be made available through intelligent security models that can flag suspicious behavior and notify security teams for appropriate action. .
5. Connect Digital Twins (Digital Copies) to Edge (Border)
The term digital twin refers to physically accurate, perfectly synchronized virtual models of real-world assets, processes, or environments. Last year, NVIDIA partnered with Siemens to support industrial metaverse use cases, helping customers accelerate adoption of industrial automation technologies. Leading companies expanding into manufacturing, retail, consumer packaged goods and telecommunications, such as BMW, Lowe’s, PepsiCo and Heavy.AI, have also begun building operational digital replicas that enable them Simulate and optimize your production environment.
What connects digital twin to the real world and edge computing is the explosion of IoT sensors and data that is driving both of these trends. In 2023, we will see organizations increasingly connecting live data from the physical environment with their virtual simulations. They will move away from simulations based on historical data towards a live digital environment – a true digital replica.
By connecting live data from the real world to their digital counterpart, organizations can better understand their environment in real time, allowing them to make faster and more informed decisions. . While it is still early days, it is expected to see strong growth in this space next year for ecosystem vendors and customer adoption.
2023 – the year of Edge AI
While the 2023 economic environment remains uncertain, Edge AI will certainly be an investment area for organizations looking to drive automation and increase efficiency. Many of the trends we saw last year continue to accelerate with a new focus on initiatives that help drive sales, reduce costs, improve customer satisfaction, and improve operational efficiency. motion.
Source: NVIDIA Blog