一种随机布设的声源定位温度补偿算法研究

One Temperature Compensation Acoustic Localization Algorithm Based on Arbitrary Layout Array

基于达时延差(TDOA)的声源定位技术具有定位精度高、实时性好、运算量小等特点,因此该技术已得到较为普遍地应用。但是针对基于TDOA的声源定位算法,声速的测量精度往往对定位精度有着较大的影响。由于工程上的声速由温度作为主要影响因素的经验公式得到,所以温度对声源定位的精度影响成为研究的重点。针对一种随机布设的声源定位算法,经研究发现,在温度的设定值接近常温变化时(25℃左右),定位算法的误差均值较小,且误差均值随温度变化单调变化。通过4组实验进行对比分析,相较于未加补偿算法,引入温补算法后的解算结果x坐标、y坐标及z坐标的误差均值分别小了0.23 m、0.25 m及0.18 m,引入温度补偿后算法的定位精度得到较大的改善。

Due to the advantages of small computational complexity, high positioning accuracy, and good real-time performance, the sound source localization algorithm based on Time-Difference-of-Arrival(TDOA)is widely used. The algorithm’s positioning accuracy is deeply affected by the measurement accuracy of the sound velocity. Since the sound velocity can be obtained by the empirical formula with temperature as the main influencing factor in engineering applications. The influence of temperature on the algorithm is the focus of our research. It is found that the regional mean errors of the acoustic localization algorithm based on arbitrary layout array are smaller than others when the experiment temperature changes around 25 ℃. And the regional mean errors change monotonically with the temperature. The 4 sets of comparative experiments were done. The mean errors of the x-coordinate, y-coordinate and z-coordinate using the temperature compensation algorithm are smaller respectively by 0.23 m, 0.25 m and 0.18 m than original result. The experimental results show that the optimized algorithm’s localization precision is better than the original algorithm.