Review of Multi-Sensor Data Fusion for Target Tracking
TANG Shengjing1， WANG Taiyan1， ZHAO Ganglian2，
GUO Jie1， LI Jiali3， YIN Hang4
1. School of Aerospace Engineering， Beijing Institute of Technology，Beijing 100081， China；2. Unit 93184 of PLA， Beijing 100076， China；3. Beijing Institute of Control and Electronic Technology， Beijing 100038， China；
4. Shanghai Electro-Mechanical Engineering Institute， Shanghai 201109， China
Abstract：With the increasing complexity of target-tracking scenarios, single-sensor detection systems cannot satisfy the practical requirements of target tracking. Multi-sensor data fusion technology has thus become an important solution for target tracking in complex scenarios. This paper reviewed the basic principles and research status of two key processes in multi-sensor data fusion: data association and estimating fusion. The relevant research on applying artificial intelligence technologies such as neural networks and reinforcement learning in multi-sensor data fusion was introduced. Finally, the development of multi-sensordata fusion methods in the field of target tracking was prospected.
