Year: 2024
Type: article
Abstract: Sensor fusion technology plays a crucial role in autonomous driving and intelligent transportation systems by integrating data from multiple sensors to achieve more accurate, reliable, and comprehensive environmental perception. Radar and cameras are commonly used sensors, each with its own advantages and limitations. Radar detects the position and velocity of objects through electromagnetic waves, offering all-weather operational capability but with low spatial resolution. Cameras capture high-resolution images using light signals, suitable for detailed object recognition, but their performance is significantly affected by lighting conditions. Sensor fusion technology can combine the strengths of radar and cameras, compensating for the shortcomings of individual sensors, and enhancing the accuracy and robustness of environmental perception. Data-level fusion processes and combines raw data to retain maximum information, improving the system's perception accuracy. Feature-level fusion extracts and combines features to reduce data volume and improve efficiency. Decision-level fusion independently processes sensor data to generate decisions, which are then fused together. Despite the significant advantages of sensor fusion technology, practical applications still face challenges such as data synchronization and calibration, uncertainty and noise, computational complexity, algorithm robustness, data transmission bandwidth limitations, and standardization issues. The forthcoming trend involves the convergence of deep learning algorithms with sensor fusion, multi-modal data fusion, online calibration technology, and the proliferation of low-cost, high-performance sensors, which will be key areas for technological breakthroughs. Continuous technological advancements and innovations will further enhance the safety and reliability of autonomous driving and intelligent transportation systems, bringing revolutionary changes to the transportation sector.
DOI: 10.54097/srbaft85
Author: Ran Tao
Institution: Sanjiang University
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4405433319