Researchers at the University of South Florida, led by Arman Sargolzaei, assistant professor in the Department of Mechanical and Aerospace Engineering and director of USF’s Resilient, Autonomous, Networked Control Systems Lab, have established a collaborative partnership with COAST Autonomous to advance research into autonomous vehicle technology. This initiative centres on the development of a high-fidelity digital twin simulation platform, facilitating a vehicle-in-the-loop approach for rigorous safety and performance validation of autonomous systems prior to real-world deployment. Sargolzaei’s work, supported by a National Science Foundation CAREER Award, has yielded a systematic framework for testing and verifying the safety and security of connected and autonomous vehicles, which will be implemented and validated through this partnership, leveraging COAST Autonomous’ expertise in autonomous systems across commercial, industrial, and governmental sectors. The collaboration extends beyond purely technological advancement to encompass investigations into community perception and interaction, addressing societal challenges alongside safety, reliability, and trustworthiness considerations within autonomous systems.
USF and COAST Autonomous Forge Research Alliance
The University of South Florida (USF) has formalised a research alliance with COAST Autonomous, a specialist in autonomous mobility solutions, to advance the development and validation of self-driving vehicle technology and to cultivate a skilled engineering workforce. This collaboration unites COAST’s practical expertise in deploying autonomous systems across diverse operational environments – including airports, seaports, rail networks, and logistics facilities – with the research capabilities of the USF College of Engineering. Central to this partnership is the development of a high-fidelity digital twin simulation platform at USF, designed to model and rigorously test autonomous systems under conditions mirroring real-world complexity. A digital twin, in this context, is a virtual replica of a physical system – in this case, an autonomous vehicle and its operating environment – allowing researchers to conduct extensive simulations and assess performance without the risks associated with physical testing.
Arman Sargolzaei, Assistant Professor in the Department of Mechanical and Aerospace Engineering at USF and Director of the Resilient, Autonomous, Networked Control Systems Lab, is leading the USF component of the research. His work, supported by a National Science Foundation CAREER Award, focuses on a systematic framework for the testing and verification of the safety and security of connected and autonomous vehicles. This framework will be implemented and validated through the partnership with COAST, providing a crucial bridge between theoretical development and practical application. The vehicle-in-the-loop approach, where simulated environments interact with real vehicle control systems, will be instrumental in assessing system robustness and identifying potential failure modes.
The alliance extends beyond purely technological advancements, addressing the critical need to build public trust in autonomous systems. Sargolzaei states that the effort extends “beyond technological advancement to encompass the building of trust in autonomous systems,” recognising that societal acceptance is paramount for widespread adoption. Furthermore, the collaboration will provide experiential learning opportunities for USF students, offering practical experience in a rapidly evolving field and contributing to the development of a skilled workforce capable of addressing the challenges and opportunities presented by autonomous vehicle technology. The focus on Autonomous vehicle safety is a key driver of the research, aiming to ensure the reliability and trustworthiness of these complex systems before deployment in public spaces.
The core of the research lies in the implementation of a systematic framework, funded in part by a National Science Foundation CAREER Award, designed to rigorously test and verify the safety and security of connected and autonomous vehicles. This framework incorporates a multi-layered approach to validation, encompassing software-in-the-loop (SIL), hardware-in-the-loop (HIL), and vehicle-in-the-loop (VIL) testing methodologies. VIL testing, facilitated by the partnership with COAST Autonomous, allows for the integration of real vehicle control systems within the simulated environment, enabling the assessment of system behaviour under realistic conditions and the identification of potential failure modes that may not be apparent in purely simulated testing.
The digital twin platform is designed to model not only the vehicle’s physical characteristics and control systems but also the surrounding environment, including road geometry, traffic patterns, and the behaviour of other road users. This necessitates the incorporation of probabilistic models to account for uncertainties in sensor data and the unpredictable nature of real-world driving scenarios. Furthermore, the research team is investigating the use of formal verification techniques to mathematically prove the correctness of critical software components, thereby enhancing the overall safety and reliability of the autonomous system. A key focus is on ensuring resilience against adversarial attacks, such as sensor spoofing or jamming, which could compromise the vehicle’s perception and control capabilities. The partnership aims to address the crucial need for establishing quantifiable metrics and standards for Autonomous vehicle safety, facilitating regulatory approval and public acceptance of this transformative technology.
Digital Twin Simulation to Enhance Safety Validation
The University of South Florida, in collaboration with COAST Autonomous, is developing a high-fidelity digital twin simulation platform to substantially enhance the validation of autonomous vehicle safety. This initiative, spearheaded by Arman Sargolzaei, Assistant Professor in the Department of Mechanical and Aerospace Engineering and Director of USF’s Resilient, Autonomous, Networked Control Systems Lab, leverages a systematic framework previously established through National Science Foundation CAREER Award funding. The digital twin is conceived not merely as a replica of the vehicle, but as a dynamic, probabilistic model encompassing the vehicle’s physical characteristics, control systems, and a realistically rendered operational environment. This environment includes detailed road geometry, nuanced traffic patterns, and the behavioural modelling of other road users – all crucial for comprehensive testing.
The platform’s architecture facilitates a vehicle-in-the-loop (VIL) methodology, integrating real vehicle control systems within the simulated environment. This allows researchers to assess system behaviour under conditions mirroring real-world complexity and to identify potential failure modes that may remain undetected through software-in-the-loop (SIL) or hardware-in-the-loop (HIL) testing alone. Crucially, the digital twin incorporates probabilistic models to address inherent uncertainties in sensor data – a fundamental challenge in autonomous navigation. These models account for the stochastic nature of real-world driving scenarios, including variations in lighting, weather, and the unpredictable actions of pedestrians and other vehicles.
Beyond environmental modelling, the research team is actively investigating formal verification techniques. These mathematical methods aim to rigorously prove the correctness of critical software components, thereby bolstering the overall safety and reliability of the autonomous system. A significant focus is placed on ensuring resilience against adversarial attacks, such as sensor spoofing or jamming, which could compromise the vehicle’s perception and control capabilities. The partnership explicitly addresses the need for establishing quantifiable metrics and standardised protocols for Autonomous vehicle safety, a prerequisite for securing regulatory approval and fostering public trust in this rapidly evolving technology. The ultimate objective is to create a robust validation framework capable of demonstrably improving the safety and trustworthiness of autonomous vehicles before widespread deployment.
Workforce Development and Trust Building
The University of South Florida’s collaboration with COAST Autonomous extends beyond technological innovation to encompass critical workforce development and the establishment of public trust in autonomous systems. Arman Sargolzaei, Assistant Professor in the Department of Mechanical and Aerospace Engineering and Director of USF’s Resilient, Autonomous, Networked Control Systems Lab, emphasizes that the initiative aims to build confidence in these complex technologies, stating the effort “extends beyond technological advancement to encompass the building of trust in autonomous systems.” This commitment is manifested through experiential learning opportunities for students, providing practical engagement in a rapidly evolving field and cultivating a skilled workforce prepared to address the challenges of autonomous vehicle deployment.
Sargolzaei’s research, funded by a National Science Foundation CAREER Award, has yielded a systematic framework for the testing and verification of connected and autonomous vehicle safety and security. The partnership with COAST Autonomous provides a unique environment for implementing and validating this framework in practical settings, bridging the gap between theoretical research and real-world application. This validation is crucial, as it moves beyond isolated component testing to encompass system-level performance and resilience under complex operational conditions. The collaborative effort is designed to address not only the technical hurdles of autonomous vehicle integration – encompassing issues of reliability, performance, and security – but also the societal implications, including community perception and interaction with these novel technologies. Establishing quantifiable metrics and standardised protocols for Autonomous vehicle safety is a central objective, essential for securing regulatory approval and fostering public acceptance. The partnership recognises that technological advancement alone is insufficient; building trust requires demonstrable safety and a workforce capable of maintaining and improving these systems throughout their lifecycle.
More information
External Link: Click Here For More
