Issue link: https://resources.mouser.com/i/1481902
STMicroelectronics 2022 29 was designed to minimize the development time while keeping high levels of flexibility to enable engineers to develop numerous applications. The FSMs are designed to use the data from the integrated sensor components embedded in sensor modules and from external sensors that can be connected to the sensor module. As part of this embedded innovation, dedicated integrated logic in sensors is configured to generate interrupt signals that are triggered by user-defined patterns. The highly flexible embedded FSMs make the latest sensors a perfect choice for the implementation of various algorithms with negligible current consumption when compared to running these algorithms in an external microcontroller. Embedded FSM, MLC, and advanced AI running in an embedded processing unit in the sensor modules is the subject of this article. The objective here is to provide engineers with a basic description of these advancements to help them select the right option for their developments. Definition of an Intelligent Sensor MEMS-based sensors can fit into three categories: Autonomous, smart, and intelligent sensors. Figure 1 illustrates a high-level block diagram of these sensors. When powered up, an autonomous sensor (Fig. 1a) that consists of a sensing element and a set of embedded electronic logic provides raw sensor data. A smart sensor (Fig.1b) is a more technologically advanced device that has certain embedded and usually hard-coded algorithms that can implement applications within the sensor. An intelligent sensor (Fig. 1c), on the other hand, has a far more advanced architecture with embedded learning and data- processing capabilities. In recent years, AI and ML have been deployed to improve efficiency, reduce cost, and increase productivity. The new generation of intelligent sensors are equipped with machine-learning/deep- learning capabilities to offer engineers the option to perform critical data processing and analysis at the deep edge of an AI-based solution. For example, this flexibility offers developers the choice to collect data through sensors and process and analyze them without any delay and latency to monitor processes, products, and assets and to gain reliable knowledge so that their applications can make the right decisions at the right time. " Designers are increasingly moving Artificial Intelligence (AI) from the center of their designs to the edge for a few important reasons. Figure 1 c. Intelligent Sensor b. Smart Sensor a. Autonomous Sensor Output: Sensor Raw Data Input: Input: Sensing Element + Electronics Output: Conditioned, compensated Sensing element + Electronics + Algorithms Input: Output: Value..., Sensing element + Electronics + Processing/ analysis Figure 1: High-level block diagram of a) an autonomous sensor, b) a smart sensor, and c) an intelligent sensor. (Source: STMicroelectronics)