Towards RobustArtificial Intelligence Approaches for Robotic Mission Planning and Execution
A PhD Proposal for a Collaboration between the University of Pisa and NATO STO Centre for Maritime Research and Experimentation
Robotic technology has advanced significantly in recent years and has found applications in various domains, including the maritime sector. The deployment of autonomous underwater vehicles (AUVs) and other types of robots in this environment is challenging due to the complexity of the underwater environment and the need for robust decision-making in real-time, often requiring fast mission re-planning.
In robotic mission planning and execution, the synergetic integration of different artificial intelligence (AI) techniques, ranging from symbolic AI rule-based strategies to data-driven machine learning and reinforcement learning (RL) methods, has the potential to improve the capabilities of these systems, making them more autonomous and flexible in accomplishing a variety of tasks.
Additionally, environmental constraints, such as the physics of acoustic propagation that impact the robot's perception and communication, must be considered to enhance the advanced decision-making and adaptability of the methods in real-world scenarios.
The proposed PhD project aims to advance the state of the art in the field of autonomy and AI for robotic mission planning and execution. The goal is to develop mission planning and execution algorithms that provide strong and formal guarantees on stability and performance, while maintaining state of the art performance on relevant tasks. The PhD candidate will work closely with researchers from the University of Pisa and the NATO STO Centre for Maritime Research and Experimentation to design, implement, and evaluate novel planning and mission execution approaches for a network of marine robots.
Existing methods have achieved state of the art performance in several tasks, but they often lack strong and formal guarantees on stability and performance. The proposed PhD project will address this limitation by exploring novel approaches which provide robustness guarantees, and by combining them with more traditional methods will investigate more effective algorithms for robotic mission planning and execution.
The PhD candidate will have the opportunity to work with a diverse team of experts from both academia and industry, including experts in robotics, autonomy, machine learning, and real-time software design. The proposed project is interdisciplinary and will leverage the strengths of both organizations, providing the candidate with unique training opportunities and access to state-of-the-art equipment and facilities.
This PhD project has the potential to have a significant impact in the field of autonomous machines, providing robots with new and effective solutions for autonomous decision-making in the maritime domain. The results of the project will be disseminated through high-quality publications in top-tier conferences and journals, and the candidate will have the opportunity to present his/her work at international conferences and workshops.
The successful candidate will hold a Master's degree in computer science, electrical engineering, or a related field, with a strong background in robotics, autonomy, machine learning and programming. He/she should also have good communication and teamwork skills, and a strong motivation to pursue a PhD degree.
