Team-based planning and execution control for multi-domain autonomous vehicles
It took only a few decades for robotic underwater, surface and air vehicles to revolutionize ocean exploration. But this is just the beginning. Vehicle heterogeneity and functional diversity will enable new teamed capabilities and new concepts of operation that could not have been imagined before. Teamed operations enable high-level specifications, interactions, and autonomy.
The main goal of this project is to develop an autonomy framework for teamed operations in the open ocean. This entails planning and execution control supported by AI-based systems, formal verification, and optimization techniques to develop a hierarchical planning and execution control architecture for teamed operations. This will involve the specification of team-level tasks, modular task decomposition, and dynamic allocation of vehicles to teams. The autonomy framework should provide assurances for overall system properties, leveraging static or runtime verification techniques.
The project builds on experience of LSTS-UPORTO (https://lsts.fe.up.pt/) in the development and deployment of networked vehicle systems for ocean observation (https://schmidtocean.org/cruise/exploring_fronts_with_multiple_aerial-surface-underwater-vehicles/). The proposed developments are applicable to other areas involving exploration of remote and communications challenged environments on Earth or in space.
Tasks and Responsibilities
- Familiarize yourself with current state-of-the-art in the fields of (1) teamed robotics, (2) models of computation for concurrent/distributed systems, (3) applications of maritime robots to ocean sciences, security, and defense, (4) software frameworks for networked underwater and surface autonomous vehicles, (5) planning, (6) execution control for autonomous vehicles, and (7) supervisory control.
- Develop formal specifications for teamed tasks and task decomposition, as well as for team-level constraints and properties, with focus on maritime operations of multi-domain vehicles.
- Develop a modular team-level planning and execution control framework enabling team-level specifications, modular decomposition, vehicle to team allocation, team-level task switching logic, and dynamic team composition.
- Evaluate and test the planning and execution control framework in a simulation environment.
- Deploy the planning and execution control framework on LSTS’s multi-domain vehicles for evaluation and testing in operational environments.
Prerequisites
Candidate must hold a Master’s degree (or equivalent) in any of the following fields: computer science, mechanical/electrical engineering, physics, mathematics, or related fields. The degree should have been completed in the last 5 years at most. The candidate must have strong analytical skills and be able to work at the intersection of computation and control. Prior experience in autonomous vehicles, concurrent systems theory, formal verification, and related fields will be a plus.
Hiring institution specific remarks
PhD Enrollment: PhD position at the Laboratório de Sistemas e Tecnologias Subaquática – LSTS (Underwater Systems and Technologies Lab) at FEUP from University of Porto (https://lsts.fe.up.pt/). The project involves collaboration with the DCC/FCUP from University of Porto and the DINF/FCT-UAc from the University of the Azores.
Co-Supervisors: Prof. João Sousa (LSTS/FEUP), contact: jtasso@fe.up.pt, Prof. Eduardo Marques (DCC/FCUP), Prof. Francisco Martins (DINF/FCT-UAc).
Notice of the Call (english version)
Notice of the Call (portuguese version)
We are no longer accepting applications for this scholarship. Thank you.