Adaptive mission planning and enactment for autonomous vehicle control

This project may involve an internship with industry, UK nationals only. 

This topic would investigate novel techniques for planning missions and ways in which the plan can be adapted in real-time as the mission is executed, to achieve a stated goal.

Aim

Development of planning methodologies for coordinated marine surveying using heterogeneous multi-robot systems, enabling efficient data collection and fusion of measurements to produce high-quality environmental surveys.

Objectives

Controlling a collection of unmanned autonomous vehicles poses the two key questions, listed below. Addressing these would be the key objectives to address during this research program.

  1. ‘How to best plan a mission using the tools available to achieve a stated aim’
  2. 'How to adapt to changes in real-time as the mission progresses in order to achieve the stated aim.'

Description

Marine environment surveying presents unique challenges that can be effectively addressed through the deployment of heterogeneous autonomous systems. These multi-robot teams, which can consist of unmanned underwater vehicles, surface vehicles, and aerial vehicles can significantly enhance both coverage rates and data quality. For example, multiple unmanned underwater vehicles (UUV) can survey the sea floor infrastructure for condition monitoring, or multiple unmanned air vehicles (UAVs) can survey the sea surface to identify partially submerged shipping containers for collision avoidance.

The heterogeneous nature of these systems, where individual platforms possess varying capabilities and sensor configurations, introduces complex planning and coordination challenges. This research addresses two fundamental questions in controlling such multi-robot teams: (1) how to optimize initial mission planning by effectively allocating available resources to meet mission objectives, and (2) how to develop adaptive planning strategies that can dynamically respond to changing conditions and new information during mission execution.

Research theme: 

Principal supervisor: 

Prof Ron Petrick
Heriot-Watt University, School of Mathematical & Computer Sciences
R.Petrick@hw.ac.uk