Robust Machine Vision and AI for Perception in Autonomous Robotic Systems

About the Project

These projects are open to students worldwide, but have no funding attached. Therefore, the successful applicant will be expected to fund tuition fees at the relevant level (home or international) and any applicable additional research costs. Please consider this before applying.

Autonomous robotic systems increasingly rely on machine vision and Artificial Intelligence (AI) to perceive, interpret, and interact with complex real-world environments. Visual perception underpins critical robotic capabilities such as scene understanding, obstacle avoidance, localisation, mapping, and interaction with dynamic surroundings. However, despite rapid progress in computer vision and deep learning, robust perception remains one of the most significant bottlenecks to reliable autonomy, particularly in challenging and unstructured environments.

Machine vision algorithms are typically developed and benchmarked under controlled or idealised conditions, where lighting, visibility, and environmental variability are relatively stable. In real-world robotic deployments, visual data is often affected by factors such as illumination changes, noise, motion blur, occlusions, domain shift, and environment-specific degradation. These effects can significantly degrade the performance of perception models, leading to cascading failures in downstream robotic tasks such as navigation and decision-making. Understanding and mitigating these limitations is therefore a critical research challenge for AI-driven robotics.

This PhD project will focus on the development and analysis of AI-based machine vision techniques that enable robust perception for autonomous robotic systems operating under challenging visual conditions. The research will investigate how visual degradation and domain variability influence core perception tasks, including depth estimation, 3D scene understanding, and visual feature representation. Rather than prioritising visual enhancement for human perception, the project will emphasise perception quality from a machine intelligence perspective, examining how AI models interpret and utilise visual information for autonomous reasoning and action.

A key aspect of the project will involve data-driven analysis using public datasets and simulated environments to systematically study the relationship between visual conditions and perception performance. The student will explore learning-based approaches that improve robustness to environmental variability, such as domain generalisation, representation learning, and task-aware perception models. The research may also consider the role of multi-task learning, uncertainty estimation, and confidence-aware perception in improving reliability for autonomous systems.

The project is designed to be algorithmic and software-led, enabling immediate progress without reliance on dedicated in-house robotic hardware. Experimental validation will be conducted using publicly available datasets, synthetic data, and simulation platforms commonly used in robotics and computer vision research. Where appropriate, the research direction may be adapted to align with the student’s background, interests, and funding source, ensuring flexibility while maintaining a strong core focus on AI and machine vision.

By addressing fundamental challenges in robust visual perception, this project will contribute to the development of intelligent robotic systems capable of operating more reliably in complex, uncertain, and visually degraded environments. The outcomes will be relevant across multiple application domains, including mobile robotics, autonomous systems, and AI-enabled perception technologies.

Decisions will be based on academic merit. The successful applicant should have, or expect to obtain, a UK Honours Degree at 2.1 (or equivalent) in a relevant discipline such as Robotics, Computer Science, Artificial Intelligence, Electrical or Electronic Engineering, Mechatronics, or a closely related field.

The candidate should have a solid foundation in robotics and autonomous systems, with a particular interest in robot perception and machine vision. Experience in programming for robotics or AI applications is essential, with proficiency in languages such as Python, MATLAB, or C++. Familiarity with computer vision, machine learning, or robotic perception algorithms is expected.

Experience with simulation-based robotics research, algorithm development, or data-driven analysis is highly desirable. Prior experience with physical robotic platforms is beneficial but not required, as the project is primarily focused on AI-driven perception and algorithmic development.

Strong analytical skills, motivation for independent research, and the ability to work at the interface of AI, robotics, and machine vision are essential.

Application Procedure:

Formal applications can be completed online: https://www.abdn.ac.uk/pgap/login.php.

You should apply for PhD in Engineering to ensure your application is passed to the correct team for processing.

Please clearly note the name of the lead supervisor and project titleon the application form. If you do not include these details, it may not be considered for the studentship.

Your application must include: A personal statement, an up-to-date copy of your academic CV, and clear copies of your educational certificates and transcripts.

Please note: you do not need to provide a research proposal with this application.

Informal enquiries can be made by contacting Dr A Rohan at ali.rohan@abdn.ac.uk. If you require any additional assistance in submitting your application or have any queries about the application process, please don't hesitate to contact us at researchadmissions@abdn.ac.uk