Exploring the Application of Multi-Sensor Fusion Technology in Enhancing the Performance of Autonomous Vehicles

Abstract

In recent years, single sensors have exposed many limitations in the field of autonomous driving, and various problems have arisen that make it difficult for single sensors to cope with complex environments. Multi-sensor fusion technology makes up for the shortcomings of a single sensor by integrating the advantages of different sensors. It not only improves the stability and accuracy of perception but also enhances the adaptability of the system in changing scenarios. The purpose of this paper is to analyze the characteristics of multiple sensors and their applications in autonomous driving systems, focusing on the fusion method based on camera and LiDAR and common fusion algorithms, such as the Kalman filter and neural network. This paper concludes that the multi-sensor fusion technology can effectively improve the target detection accuracy, enhance the system robustness, and show significant advantages in obstacle sensing and obstacle avoidance strategies. In addition, enhancement schemes such as optimized fusion algorithms, intelligent modal selection, and edge computing are proposed to address challenges such as real-time and computational complexity. This study provides a theoretical foundation and practical reference for the further application of multi-sensor fusion technology in autonomous driving.