Acoustic Source Tracking Using Track Before Detect Based on the Fusion with the Prominent SRP Measurement
Yao Guo, Hongyan Zhu · 2022 41st Chinese Control Conference (CCC) · 2022
Aiming at the problem that the prominent measurement is not fully utilized in the process of the track before detect (TBD) method, an improved particle filter implementation method for tracking acoustic source is proposed. The proposed method introduces the spectral peak information of the steered response power (SRP) to the process of particle evolution, and improves the speed of particle's convergence to the true position. The method has the advantages of TBD when dealing with weak source signal, and can combine the prominent measurement at the same time, which provides a solution for accurate source tracking. The experimental results show that the proposed method has better tracking performance than the traditional TBD methods under indoor conditions.