Our clients produce a wide variety of electronic devices that affect all aspects of our modern life, including transportation, healthcare, communications, etc. Our discussions tend to focus on how we can skillfully set the scene between reality and the digital world Bridge, and discuss the innovations they hope to achieve in the future, and I conclude these five macro-trends in science and technology that will have the greatest impact on business and society by 2018, based on these and other studies.
Artificial intelligence
Just as people struggled to realize the digital advantage ten years ago, customers in every market segment are fanatically trying to understand the value of artificial intelligence and machine learning to their business. As barriers such as performance and affordability continue to erode , Plus the fact that some AI-specific applications are gradually generating benefits and application-level effects in industrial environments, attention to AI applications will accelerate in 2018. For example, AI has evolved to the point where industrial robots learn and adapt without special training New environment or stage of unfamiliar objects.
With innovations in low-power signal processing, the AI at the edge nodes will start to transform from novelty to conventional technology, while using more contextual data and information to drive smarter system partitioning between the edge and the cloud, At the same time, AI applications that compete with human intelligence will continue to be dominated by university research.
Autonomous / intelligent machines
Autonomous systems for cars, drones and robots will continue to evolve in 2018, but will be limited by some pending regulatory and technical issues, but in the coming months some of the autonomous systems-based project plans We will continue to see progress, such as the pilot deployment of driverless taxis in restricted areas, in particular long-distance transport such as trucks and trains, which will be one of the applications that have made substantial progress in the near-term unmanned drive.
Driven by the constant pursuit of productivity benefits, the drive to add intelligence to machines will also accelerate the Factory Automation / Industry 4.0 process, where, for example, advances in machine learning will significantly improve the system's ability to deliver valuable performance based on its own independent status monitoring Advice and forecast.
Ubiquitous wireless sensing network and data
The combination of advanced materials, enhanced functionality, and MEMS combine to make sensor size and cost a breakthrough, making wireless sensor networks ubiquitous possible Deployment through wireless mesh networks in IoT and industrial applications To add sensing capabilities to existing systems without extensive rewiring, but end-to-end security from sensors to the cloud will be a fundamental requirement for industrial users to deploy industrial IoT on a large scale.
The drive to make products and systems smarter will also increase the need to manage and analyze growing data flows as the data load continues to increase, data centers will require higher processing performance, and advanced power management Innovation to mitigate the risk of high-temperature in data center systems, and we will also gradually see edge nodes consolidate more intelligence and begin to categorize and process data flows.
Human-machine interface
As augmented reality and virtual reality ecosystems thrive and encourage more innovation, mixed reality systems will continue to emerge and gain popularity, as their commercial AR / VR systems accelerate, their costs will decline and their reach will extend to Off-site diagnostics and repair in areas such as industry.
In addition, the voice user interface has been well-received by people, but this technology still faces some limitations, especially in noisy environments.Gartner predicts that 30% of 2018 technology interactions will be 'talked' to smart machines, This means that technology and service providers need to invest now to improve the currently limited voice interface.
Heterogeneous architecture manufacturing
With the soaring cost of deep sub-micron technology development and Moore's Law being challenged by increasing technology and costs, the heterogeneous architecture consolidation of multiple technologies in a single package, single-layer laminate, and even on a single silicon substrate will increase New business models to capitalize on heterogeneous fabrications will emerge, enabling reorganization and innovation for small-scale semiconductor vendors that can not afford to invest in state-of-the-art IC lithography, and for larger, larger suppliers , The signal processing algorithm will be integrated into the chip will help to enhance the value of its program.
in conclusion
As the saying goes, the best way to predict the future is to create the future, when semiconductor innovations will form the basis of these emerging applications, and analogies will become more in the 'data hungry' world Importantly, ADI will make every effort to make these predictions a reality by 2018.