With the advantages of high velocity measurement accuracy and fast dynamic response, the laser Doppler velocimeter (LDV) is expected to replace the odometer to be combined with a strapdown inertial navigation system (SINS) to form a higher precision integrated navigation system. However, LDV scale factor error and misalignment angles between LDV and inertial measurement unit will affect the accuracy of navigation. Considering that not all global navigation satellite system (GNSS) receivers can directly provide velocity information and current mainstream calibration methods are sensitive to the measurement noise and outliers of velocity and position information, a robust calibration method aided by GNSS is proposed in this paper, which is based on position observation. Different from current popular calibration methods, the attitude information of the GNSS/SINS integrated navigation system obtained by an adaptive Kalman filter is used to construct the observation vector together with LDV velocity outputs and GNSS position outputs in this method. The LDV scale factor error and the misalignment angle are determined by the ratio of two observation vector modulus and the Davenport’s q-method method, respectively. The accuracy and robustness of the calibration method are verified by one vehicle test with normal GNSS signals and one vehicle test with GNSS signals with outliers. And the horizontal position error of dead reckoning of the calibrated LDV/SINS integrated system are less than 0.0314% and 0.1033% of the mileage, respectively.

In the integrated navigation system, which consists of a laser Doppler velocimeter (LDV) and strapdown inertial navigation system, it is necessary to calibrate the LDV. We propose an online calibration method for a LDV based on the strapdown inertial navigation system. This system uses highly accurate output information in the early stage of the inertial navigation system to carry out coarse calibration for the velocimeter and then employs a Kalman filter to correct the results of coarse calibration to achieve high precision calibration for the velocimeter. Experimental results show that the relative distance error of dead reckoning is less than 0.02%, indicating that the proposed online calibration method can achieve high-precision calibration for the velocimeter.

In the autonomous navigation of drones, improving positioning accuracy is of significant importance to obtain highly accurate information on flight velocity. Traditional microwave and acoustic velocity measurement methods have the disadvantages of poor precision and susceptibility to interference. In this study, an unmanned aerial vehicle (UAV)-mounted two-dimensional laser Doppler velocimeter was developed and investigated, and a relevant drone flight navigation and positioning experiment was carried out. The UAV-mounted two-dimensional laser Doppler velocimeter (LDV) prototype developed in this study applies a scheme of dual-beam measurement light, sharing a focusing lens group. After process integration, the performance of the prototype was measured. It shows that a velocity measurement effect with a high signal-to-noise ratio can be achieved by using two measurement probe beams within a working distance range of 40 m–60 m. In the flight experiment, the flight trajectory calculated using the LDV-measured velocity data was compared with the global navigation satellite system (GNSS)-recorded trajectory. The result shows that LDV can achieve an odometer accuracy of 4.8‰. This study has validated the feasibility of the laser Doppler velocimeter in drone navigation and positioning, providing a novel method for reliable and high-precision velocity measurement in autonomous drone navigation.

In order to reduce the requirement of system bandwidth of Laser Doppler Velocimeter (LDV), a Dual-Doppler signal mixing LDV is proposed in this paper. By transmitting two beams to the moving surface, two Doppler signals are acquired and subsequently mixed to obtain a difference frequency signal. The measured speed can be calculated based on the frequency of this difference frequency signal. This novel structure significantly reduces the bandwidth requirements on the system, which can be further diminished by minimizing the angle between the two beams of the emitted light. Moreover, it exhibits enhanced robustness against variations in launch angle and enables defocusing measurements.

When the amplitude of the laser Doppler signal is less than that of the noise amplitude, the speed information cannot be identified. In this paper, the recognition of weak Doppler signals is accomplished by superimposing multiple-frame Doppler signals in the frequency domain and eliminating the base noise. Meanwhile, the relative prominence is defined to assess the quality of Doppler signals. The theoretical analysis and experimental results demonstrate that the relative prominence can be enhanced by more than two orders of magnitude through the stacking of 10,000 frames of Doppler signals. Moreover, this paper establishes a foundation for determining the threshold of relative prominence in practical measurements. When the relative prominence at a specific point on the spectrum is 3, there is at least a 99.08% probability that this point corresponds to the target signal.

Quality factor is an important evaluation index of Laser Doppler Velocimeter (LDV). In order to study the effect of spot size on quality factor, the effects of signal light intensity and transit time on the quality factor are discussed theoretically. In addition, the distribution of quality factor of LDV was measured and examined. Theoretical analysis and experimental results demonstrate that: Due to the symmetry of Gaussian beam, the distribution of quality factor is symmetric. Furthermore, the light intensity of Doppler signal at the waist spot of Gaussian beam is the largest while the transit time of moving particles diminishes due to the decrease of spot diameter, causing spectrum leakage. Under the influence of both Doppler signal intensity and transit time, the distribution of quality factor appears concave phenomenon.