FMCW LiDAR with an FM nonlinear kernel function for dynamic-distance measurement

Phase tracking using a Kalman filter based on probability density distribution in frequency-scanning interferometry

Frequency-scanning interferometry (FSI) utilizing external cavity diode lasers (ECDL) stands out as a potent technique for absolute distance measurement. Nevertheless, the inherent scanning nonlinearity of ECDL and phase noise pose a challenge, as it can compromise the accuracy of phase extraction from interference signals, thereby reducing the measurement accuracy of FSI. In this study, we propose a composite algorithm aimed at mitigating non-orthogonal errors by integrating the least-squares and Heydemann correction technique. Furthermore, we employ Kalman filtering for precise phase tracking. We introduce a parameter selection strategy based on the statistical distribution of instantaneous frequency to achieve the fusion estimation of phase observation values and theoretical models, which starts a new perspective for the application of multi-dimensional data fusion in FSI measurement. Through simulation and experimental validation, the efficacy of this approach is confirmed. The experimental results show promising outcomes: with an average phase error of 0.12%, a standard deviation of less than 1.7 µm in absolute distance measurement, and an average positioning accuracy error of 0.29 µm.

High-speed nonlinear frequency sweeping signal distance extraction algorithm based on the table lookup method

In the domain of frequency sweeping interferometry, the accurate extraction of distance information from nonlinear frequency scanning signals holds paramount significance in ensuring meticulous measurements of high precision. This paper presents a novel, to the best of our knowledge, high-speed distance extraction algorithm based on the table lookup method and validates its feasibility through theoretical models, simulations, and practical experiments. The proposed algorithm achieves comparable accuracy to traditional methods involving resampling and Hilbert transform. However, it outperforms them in robustness against noise and variations in sampling points. This method can accurately process signals sampled even below the Nyquist sampling rate. The simplicity and computational efficiency of the proposed approach make it suitable for various nonlinear sampling applications, promising broad applicability in scientific and engineering contexts.

Laser swept-frequency interferometry with non-uniform fast Fourier transform and a linear regression frequency measurement method

The laser swept-frequency interferometric ranging method is commonly used in the field of large-scale, high-precision, and non-cooperative measurements. However, this method requires the laser chirp curve to be a stable straight line. Nonlinearities in the chirp can cause broadening of the target spectrum, which affects the accuracy of the frequency extraction of the beat signal, resulting in increased ranging error. Herein, a linear regression laser swept-frequency interferometry method based on the non-uniform fast Fourier transform is proposed, which effectively suppresses the influence of frequency modulation nonlinearity on ranging accuracy.

Real-time high-precision FMCW laser range extraction method based on a hardware multiplier array

Frequency-modulated continuous wave (FMCW) laser interferometry is an ideal large-scale absolute distance measurement method. It has advantages of high precision and noncooperative target measurement capability, with no blind spot for ranging. To meet the requirements of high-precision, high-speed 3D topography measurement technologies, a faster measurement speed of FMCW LiDAR at each measurement point is required. To solve the shortcomings of the existing technology, a real-time high-precision hardware solution method (including but not limited to FPGA and GPU) for lidar beat frequency signals is provided here based on hardware multiplier arrays to reduce lidar beat frequency signal processing time and to save energy and resource consumption during processing. A high-speed FPGA architecture was also designed for the frequency-modulated continuous wave lidar range extraction algorithm. The whole algorithm was designed and implemented in real time based on the principle of full-pipelines and parallelism. The results show that the processing speed of the FPGA system is faster than that of current top-performing software implementations.