Intelligence at the Edge Powers Autonomous Factories

alan000345

Intelligence at the Edge Powers Autonomous Factories [Copy link]

From traditional industrial robotic systems to today’s latest collaborative robots, all types of robots rely on sensors that can generate and process large amounts of highly variable data. This data can be used to enable autonomous robots that can make real-time decisions, enabling smarter event management while maintaining productivity in dynamic real-world environments, as shown in Figure 1.

查看详情

Figure 1: Millimeter wave (mmWave) sensing helps monitor the area around a machine and enables real-time event management

How TI mmWave sensors enable advanced intelligence in factories

Millimeter wave (mmWave) sensors from Texas Instruments (TI) are able to process data on-chip using an integrated processor to enable real-time decision making. This integration enables smaller designs compared to some light- or vision-based sensors. In addition, the ability to detect multiple objects and process data using only a single sensor reduces overall system costs.

Another important consideration for factory environments is immunity to environmental conditions such as dust, smoke, and variable lighting. mmWave sensors can operate in any of these conditions - and are mounted behind the plastic of the housing - without the need for external lenses, apertures, or sensor surfaces. All of these attributes make mmWave sensors perform well in industrial sensing applications.

Intelligent edge processing allows factory machines and robots to interact with people and reduce accidents. For example, TI mmWave sensors can be configured to monitor specific areas of interest around machinery, define prohibited areas, and warn people in the area. These areas can be partitioned so that sensors can react accordingly based on area occupancy or human proximity. Figure 2 demonstrates this capability, where green areas indicate safety; yellow areas indicate warnings, and red areas indicate danger, indicating proximity to machinery.

查看详情

Figure 2: A person slowly walks past a machine with a danger signal flashing 1 meter away

TI mmWave sensors can accurately measure not only the distance of objects in the field of view, but also the relative speed of any obstacles. This enables the robot to take more predictive actions, such as stopping the machine, based on how quickly an object approaches the sensor. Figure 3 shows how quickly a machine can trigger a danger zone warning based on the speed of a person approaching the machine.

查看详情

Figure 3: When a person walks slowly, the danger sign is triggered at 1 meter (a); when a person walks fast, the danger sign is triggered at 2 meters (b)

To increase productivity, you want the machine to avoid stopping due to false triggers. The example in Figure 4 shows how the integrated tracking algorithm enables the sensor to accurately determine the direction of a person. When a person leaves the machine, it does not turn on a warning signal or take other actions.

查看详情

Figure 4: The sensor does not indicate a danger signal because a person is leaving the machine

Simplify robot design and speed up development

To simplify the design of robotic systems and reduce development time, visit our Area Scanner demo. It features a new static object detection algorithm that detects static objects that may remain in the robot's "no-go zone", such as boxes, carts, or other equipment, while ignoring permanent static objects, such as pillars that may have become part of the scene. The demo runs on the TI mmWave IWR6843ISK, IWR6843ISK-ODS, and IWR6843AoP evaluation modules (EVMs).

Using our antenna-on-chip (AOP) sensors, which feature a radio frequency (RF) antenna etched into the chip’s package, can help you design a smaller system by saving 75% board area compared to other radar technologies, as shown in Figure 5.

查看详情

Figure 5: Using TI mmWave packaged antenna sensors, the sensor size can be reduced by up to 75%.