Grant: Discovery Grants Program - Individual
The University of Western Ontario
$36,000.00
Intelligent Driving Autonomy: A Multi-Faceted Approach with Fairness-Aware Monitoring and Attention-Based Perception for Contextually Adaptive Handover
FY 2024-25
Summary
A vast majority of severe road accidents in Canada - over 90% - arise from human errors caused by distraction, fatigue, or impairment. Modern technologies, especially artificial intelligence (AI), are presenting new ways to keep our roads safer. Think of AI as the brain behind self-driving cars, helping them drive for long stretches without human intervention. However, there are still questions about whether we can trust these AI systems completely. In partial/conditional automation, the car might need to give control back to the driver due to technical challenges or legal mandates. This research proposal aims to design an innovative system that can help with the control transition between the driver and self-driving car, making sure it is safe based on the context - such as "how busy the road is" or "how alert the driver seems".
The long-term goal of the proposed research program is to advance driving autonomy, improve accessibility, and elevate road safety. My research revolves around three themes: A) Driver monitoring: Designing a real-time system that can check if the driver has a normal physical/mental state and is attentive and ready to control the vehicle. A cornerstone of this endeavor is to eliminate potential biases in this monitoring technology, ensuring every driver is perceived through a lens of equity. B) Enhancing how cars perceive surroundings: By harnessing insights from human attention patterns, specifically their eye movements, we aim to amplify the reliability of the automated driving system by teaching the vehicle to focus on important objects the way humans concentrate. C) Intelligent control transition: Designing a control handover system that seamlessly navigates the optimal switch between the human driver and automated driving system. Depending on various driving situations, this system will decide the best time for either the car or the driver to control the vehicle, using a special kind of learning, called reinforcement learning (RL).
With the help of advanced machine learning methodologies, this research program will address the outlined objectives and train highly qualified personnel. Ultimately, my goal is to improve road safety, using an optimal balance of driver state and smart car technology.
Program
Discovery Grants Program - Individual
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
RGPIN-2024-05712
Name nom
Nikan, Soodeh
Department département
Electrical and Computer Engineering
Organization id
36
Institution établissement
The University of Western Ontario
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
36000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1510
Committee name en
Electrical and Computer Engineering
Committee name fr
Génie électrique et informatique
Area of application code
700
Area of application group en
Transportation systems and services
Area of application group fr
Services et systèmes de transport
Area of application en
Transportation systems and services
Area of application fr
Services et systèmes de transport
Research subject code
2800
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Artificial intelligence (computer vision, use2603)
Research subject fr
Intelligence artificielle (vision artificielle, utiliser 2603)
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Intelligent Driving Autonomy: A Multi-Faceted Approach with Fairness-Aware Monitoring and Attention-Based Perception for Contextually Adaptive Handover
Keywords
computer vision; autonomous vehicle; scene perception; deep learning; salient object detection; reinforcement learning; knowledge distilation; automated driving; driver monitoring; artificial intelligence
Application summary
A vast majority of severe road accidents in Canada - over 90% - arise from human errors caused by distraction, fatigue, or impairment. Modern technologies, especially artificial intelligence (AI), are presenting new ways to keep our roads safer. Think of AI as the brain behind self-driving cars, helping them drive for long stretches without human intervention. However, there are still questions about whether we can trust these AI systems completely. In partial/conditional automation, the car might need to give control back to the driver due to technical challenges or legal mandates. This research proposal aims to design an innovative system that can help with the control transition between the driver and self-driving car, making sure it is safe based on the context - such as "how busy the road is" or "how alert the driver seems".
The long-term goal of the proposed research program is to advance driving autonomy, improve accessibility, and elevate road safety. My research revolves around three themes: A) Driver monitoring: Designing a real-time system that can check if the driver has a normal physical/mental state and is attentive and ready to control the vehicle. A cornerstone of this endeavor is to eliminate potential biases in this monitoring technology, ensuring every driver is perceived through a lens of equity. B) Enhancing how cars perceive surroundings: By harnessing insights from human attention patterns, specifically their eye movements, we aim to amplify the reliability of the automated driving system by teaching the vehicle to focus on important objects the way humans concentrate. C) Intelligent control transition: Designing a control handover system that seamlessly navigates the optimal switch between the human driver and automated driving system. Depending on various driving situations, this system will decide the best time for either the car or the driver to control the vehicle, using a special kind of learning, called reinforcement learning (RL).
With the help of advanced machine learning methodologies, this research program will address the outlined objectives and train highly qualified personnel. Ultimately, my goal is to improve road safety, using an optimal balance of driver state and smart car technology.
