German component supplier Bosch calls its new semiconductor fab in Dresden a smart factory with highly automated, fully connected machines and integrated processes combined with AI and internet of things (IoT) technologies for facilitating data-driven manufacturing. With machines that think for themselves and glasses with built-in cameras, maintenance work in this fab can be performed from 9,000 kilometers (about 5,592 miles) away.
Not surprisingly, Bosch sees this fab as a manifestation of Industry 4.0, which implies that production units feature smart processing capabilities and communicate without human aid, all while bridging the physical and virtual worlds.
“With the help of artificial intelligence, we will take semiconductor manufacturing to the next level,” said Volkmar Denner, chairman of the board of management at Robert Bosch GmbH.
Industry 4.0 encompasses the extensive demands of smart manufacturing in the automotive and several electronics industries. Here, Industry 4.0 solutions support automated production line supervision with sensor-based systems with real-time monitoring and maneuverability for any manufacturing process.
It is also important to note that the notion of Industry 4.0-which emerged around a decade ago-has been continuously evolving amid the evolution of embedded systems, internet connectivity and data analytics. In the Industry 4.0 realm, sensor-based devices are constantly reshaping the industrial, automotive and robotics segments; that includes industrial pressure transmitters, HVAC sensors, weight scales, factory automation devices and smart meters.
Evolution of smart manufacturing
However, before delving into the role of smart sensors in Industry 4.0, it’s worthwhile to see how the industry reached this inflection point.
It all began in the 1980s, when companies connected computers to control equipment and began using robots for automated material handling. Factory automation followed suit by automating repetitive manufacturing processes, such as material transportation.
Then came the idea of smart manufacturing, which began synergizing IoT, data-driven analytics and AI-powered technologies. If done effectively, that leads to process optimization, flexible production, shortened delivery time, manpower quality and enhanced efficiency.
But that also calls for more compute power, advanced motor-control algorithms, reliable connectivity and stable image recognition. Likewise, smart sensors are a key ingredient in the transformation of manufacturing processes. Smart sensors take analog input from the physical environment and employ specialized technologies to condition this data for use in the digital world.
Take the example of an autonomous rover for smart manufacturing, which uses navigation algorithms in conjunction with time-of-flight (ToF) sensors to move freely without causing a collision. It also employs AI and machine vision technologies to identify objects of specific colors, pick them up, and place them in the corresponding baskets using its robotic arm with field-oriented control (FOC).
Sensor building blocks: Old and new
Signal conditioners are still vital in smart sensors operating mostly on non-linear and very small signals that require specialized technologies to convert the sensor signal into a linearized output. Signal conditioners bolster sensor interfaces by providing programmable, highly accurate, wide gain and quantization functions combined with powerful, high-order digital correction and linearization algorithms.
Then there are image sensors; though not a new technology, they are playing a critical role in smart manufacturing applications by facilitating high-performance industrial inspection automation. Advanced CMOS image sensors are now a staple in 3D machine vision and spectral analysis commonly used in smart manufacturing environments.
Alongside these tried and tested sensor building blocks, new technologies are emerging to cater to more demanding applications in the smart manufacturing realm. Take, for instance, ToF sensors commonly used to capture 3D images in industrial operations. Such sensing cameras enable machine vision and predictive maintenance in robotics, Industry 4.0 and automotive plants.
Another notable technology, comprising highly miniaturized MEMS scanners, is increasingly employed for image scanning in industrial manufacturing. These devices-1D and 2D micro-scanners-feature large scan angles and high scan frequencies to create highly reliable control systems in manufacturing environments.
AI meets MEMS sensors
Miniaturized, intelligent, and networked sensors and actuators form the basis for IoT, Industry 4.0 and smart manufacturing applications employing AI features-and these smart sensors are increasingly finding a place in applications at the industrial and IoT edge.
Case in point: STMicroelectronics (STMicro) has added compute power to sensing in what it calls an intelligent sensor processing unit (ISPU). It combines a DSP suited to run AI algorithms and MEMS sensor on the same chip. The merger of sensors and AI puts electronic decision-making on the edge, while enabling smart sensors to sense, process and take actions, bridging the fusion of technology and the physical world.
STMicro is also prepping its STM32 processors for applications at the industrial and IoT edge by adding AI acceleration, as well as time-sensitive networking (TSN) support and PCIe, USB 3.0 and CAN-FD peripherals. As a result, these AI-powered processors are ready for emerging opportunities in Industry 4.0, IoT and rich user-interface applications, according to Ricardo De Sa Earp, executive VP and general manager of the General-Purpose Microcontrollers Sub-Group GM at STMicro.
The above examples show how manufacturing is evolving in an era of smart sensors and compute abundance combined with AI and neural networks. Smart sensors, which form the backbone of Industry 4.0, are increasingly used in smart factory environments. They have become a key Industry 4.0 ingredient alongside AI, cloud computing and IIoT.