Researchers are working to improve the aerodynamic efficiency of autonomous vehicles (AVs) by minimizing the drag caused by sensors like cameras and LiDAR systems mounted on the exterior. They created an automated computational system that integrates experimental design with a substitute model and an optimization algorithm, which helped refine the shapes of these sensors. After conducting simulations on both standard and optimized designs, they discovered that the optimized design led to a 3.44% reduction in overall aerodynamic drag for an AV.
With the rapid advancement of information technology and artificial intelligence, autonomous vehicles (AVs) are becoming increasingly prevalent. The technology has progressed to the point where these vehicles are now employed for logistics and low-speed public transportation services.
While significant research has been aimed at enhancing control algorithms for safety improvements, there has been less focus on boosting aerodynamic efficiency. This enhancement is vital for decreasing energy usage and extending travel distances. Consequently, issues related to aerodynamic drag are hindering the ability of self-driving vehicles to match the acceleration capabilities of conventional vehicles.
In a study published in Physics of Fluids by AIP Publishing, a team from Wuhan University of Technology in China concentrated on improving the aerodynamic performance of AVs by lessening the drag from externally mounted sensors, such as cameras and LiDAR, which are crucial for AV operation.
According to author Yiping Wang, “Sensors mounted externally substantially elevate aerodynamic drag, especially by increasing the share of interference drag within the overall aerodynamic drag.” Wang noted that understanding the interactions between sensors and how their geometric dimensions affect interference drag is important for thoroughly optimizing sensors during the design phase.
The team utilized a blend of computational techniques and experimental approaches. They set up an automated computational framework and integrated experimental design with a substitute model and optimization algorithms to enhance the designs of AV sensors. Subsequently, they conducted simulations on both the original and improved models to analyze drag reduction and the aerodynamic performance enhancements of the optimized version. They also performed wind tunnel tests to confirm the accuracy of their results.
Post-optimization, the researchers identified a 3.44% reduction in total aerodynamic drag for AVs. In comparison to the initial model, the optimized version demonstrated a 5.99% decrease in the aerodynamic drag coefficient during simulations and showed a marked improvement in aerodynamic performance under changing conditions.
The team also noted enhancements in airflow around the sensors, with reduced turbulence and better pressure distribution at the rear of the vehicle.
Wang stated, “Looking forward, our research could guide the creation of more aerodynamically efficient autonomous vehicles, allowing them to cover greater distances. This is particularly significant as the use of autonomous vehicles grows, not just in passenger transport but also in delivery and logistics sectors.”
