The military has always been the pioneer of sensor data fusion, “by using data from multiple distributed sources, we can obtain lower detection error probability and higher reliability”.

Nowadays, advanced sensor applications in smart phones and cars often use sensor fusion. However, advanced industrial applications (such as complex robots) also use the technology. Sensor fusion is actually a subcategory of data fusion, also known as multi-sensor data fusion or sensor data fusion.

For position sensing in smartphones, the combination of data from accelerometers, gyroscopes and magnetometers can provide better results than any of these sensors alone. In cars, data from radar, lidar and cameras, as well as maps and other data sources, are combined to make decisions in autonomous driving situations. In robot, data from vision system and inertial measurement unit (IMU) are combined to improve the motion of manipulator and increase the sampling rate.

The military has always been the pioneer of sensor data fusion, “by using data from multiple distributed sources, we can obtain lower detection error probability and higher reliability”.

In 1987, the JDL data fusion group of the US Department of Defense (DoD) developed the definition of data fusion, which is further improved as “multi-level and multi-faceted process, involving automatic detection, correlation, correlation, estimation, and the combination of data and information from single or multiple sources.”

Top level view of JDL data fusion process model

Technical requirements of sensor data fusion

Leave a Reply

Your email address will not be published. Required fields are marked *