MFCC-Calib: A generalized algorithm framework for calibrating LiDAR-IMU exterior orientation elements based on multi-feature control and constraint method

• Published in: Optics and laser technology Vol. 189; p. 113019
• Main Authors: Shi, Bo, Yang, Xingyi, Zhao, Kai, Ren, Hongwei, Yang, Fanlin
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2025
• Subjects: Constraint calibration method; Control calibration method; Inertial Measurement Unit (IMU); Light Detection and Ranging (LiDAR); Mobile mapping system (MMS); Multi-feature calibration method; Control calibration method; Mobile mapping system (MMS); Multi-feature calibration method; Constraint calibration method; Light Detection and Ranging (LiDAR); Inertial Measurement Unit (IMU)
• ISSN: 0030-3992
• DOI: 10.1016/j.optlastec.2025.113019

The Mobile Mapping System (MMS), comprised of Light Detection and Ranging (LiDAR), the Global Navigation Satellite System (GNSS) and the Inertial Measurement Unit (IMU), is frequently utilized for the acquisition of 3D spatial data. It is essential to note that the exterior orientation elements between LiDAR and IMU directly influence the overall quality and accuracy of the acquired data. This paper proposes a generalized algorithm for calibrating LiDAR-IMU exterior orientation elements based on the combination of multi-feature control and constraint method, which makes the calibration model more general, reduces memory usage, and enhances the computational speed. In the calibration model, we first present the construction method for a single-feature calibration model in a general sense, which is subsequently extended to develop multi-feature calibration models. Additionally, the least squares adjustment with conditions and constraints method is employed to formulate the feature calibration model, which expresses the calibration method based on feature control and constraints in a unified form. Finally, this model is versatile and can be applied to calibration tasks involving diverse observations and exterior orientation elements. In terms of algorithm implementation, the feature calibration model presented in this paper employs the summation of normals algorithm during the construction of the single-feature calibration model. This approach eliminates the need for high-dimensional matrix storage and operations, thereby enhancing computational efficiency. Furthermore, the parallel construction of multi-feature, based on the independence of different features in the normal equation construction process, significantly reduces computational time. In a dataset comprising 2 million data points of 15 features, the calibration process requires less than 3s of computational time and occupies less than 1MB of memory usage. To assess the performance of the algorithm, we analyze both time and space complexity. The time complexity is represented as O(n), being directly proportional to the size of the observation dataset and inversely proportional to the available parallel computing resources. Simultaneously, space complexity scales linearly with the number of features. These patterns are verified using empirical datasets. •A generalized framework for calibrating LiDAR-IMU exterior orientation is proposed.•The calibration framework flexibly combines multi-features with control and constraint method.•The framework based on the parallel structure of MFCC method, which improves the speed of calibration.