- Researchers devised an automated computational platform that facilitated the optimisation of sensor shapes integral to AV design.
Caption: Deformation control volumes are set for the front sensor, front-side sensor, roof sensor, and rear-side sensor, which significantly impact the aerodynamic drag coefficient. The sensor shapes can be modified by adjusting the control points on these control volumes.
The advent of information technology and artificial intelligence has heralded a new era in transportation, particularly through the development of autonomous vehicles (AVs).
The transformative innovation is not only revolutionizing logistics delivery and low-speed public transportation but also amplifying the call for improved efficiency and sustainability in vehicular technology.
Externally mounted sensors
While significant attention has been dedicated to refining control algorithms aimed at ensuring safety, an equally critical area that requires scrutiny is the aerodynamic performance of these vehicles.
Enhancing this aspect is crucial for reducing energy consumption and extending driving ranges, an objective that remains pertinent as AVs integrate further into everyday use.
Research conducted by a team from Wuhan University of Technology underscores the pressing need to address the aerodynamic drag challenges faced by AVs.
A pivotal concern arises from the externally mounted sensors—such as cameras and light detection and ranging (LiDAR) instruments—that are indispensable for the operational effectiveness of AVs.
Yiping Wang, one of the study’s authors, highlights a significant drawback of such configurations: “Externally mounted sensors significantly increase aerodynamic drag, particularly by increasing the proportion of interference drag within the total aerodynamic drag.”
A meticulous process
The observation accentuates the necessity for a holistic optimisation approach during the design phase, wherein the interactions between sensors and the geometric dimensions affecting aerodynamic drag can be meticulously addressed.
Employing a robust blend of computational and experimental methodologies, the researchers devised an automated computational platform that facilitated the optimisation of sensor shapes integral to AV design.
Through a meticulous process that integrated experimental design, substitute modeling, and optimisation algorithms, the researchers successfully reengineered the structural configurations of the AV sensors.
The efficacy of these enhancements was validated through a series of simulations comparing both the baseline and optimised models. Notably, a wind tunnel experiment corroborated the findings, demonstrating a measurable reduction in aerodynamic drag.
The results of the optimisation process are compelling; the study revealed a striking 3.44 per cent decrease in the total aerodynamic drag of the autonomous vehicle. Simulations indicated a more profound reduction of 5.99 per cent in the aerodynamic drag coefficient when juxtaposed with the base model.
Furthermore, improvements in airflow dynamics were observed, characterised by diminished turbulence around the sensors and more favourable pressure distributions at the rear of the vehicle.
These aerodynamic enhancements not only signify progress in the quest for efficiency but also align with broader sustainability goals as the world grapples with energy consumption challenges.
Looking toward the future, the implications of this research are profound.
As the adoption of autonomous vehicles becomes more prevalent across various sectors, particularly in passenger transport and logistics applications, the need for aerodynamically efficient designs is paramount.
Wang eloquently asserts that the outcomes of this study could inform future design paradigms, facilitating the construction of autonomous vehicles capable of traveling longer distances with reduced energy expenditures.