An Universal Data Fusion Method for Velocity Measurements of Multi-Sensor

In multi-sensor system, Doppler velocity can be obtained directly from sensor such as radar or sonar while velocity parameters in Cartesian coordinates can only be obtained from filtering results. In order to acquire higher velocity precision, an universal data fusion is needed. As a new location algorithm, JPDA (Joint Probability Density Algorithm) is extended to estimate target velocity when Doppler velocity and Cartesian velocity can be obtained synchronously. Firstly, two velocity probability density functions are derived from Doppler measurements and velocity in Cartesian coordinates respectively. Secondly, joint velocity probability density is constructed and JPDA is implemented. As an extension of JPDA, the new method proposed here could estimate target velocity with higher precise and there is no limitation on sensor type or account. Simulation results verify the feasibility of the method.