Thomas E. Kent

Thomas E. Kent

Lead Machine Learning Scientist

Mind Foundry



  • Artificial Intelligence
  • Decision Making
  • Path Planning
  • Multi-Agent Systems
  • Simulation
  • Optimisation


  • PhD in Aerospace Engineering, 2015

    University of Bristol

  • Master of Arts with Honours Pure Mathematics, 2011

    University of Edinburgh


Some of my previous projects


Hiearchical Decision Making in Multi-Agent Systems

TB-Phase Project

Jan 2018 – Mar 2022 University of Bristol
As part of the TB-Phase project: a 5 year EPSRC Prosperity Partnership researching new engineering design for hybrid autonomous systems. My research explores the use of Multi-Agent systems for tasking, routing and surveillance problems. Using Evolutionary Algorithms and Reinforcement Learning to develop local single-agent policies capable of scaling to multi-agent scenarios.

Decision Making for Driverless cars

Venturer Project

Jan 2015 – Apr 2018 Bristol Robotics Laboratory
This project employed a number of my research interests, areas of control, optimisation, simulation and mathematical modelling within a robotics setting. I was tasked with the design, implementation and testing of a number of motion planning and decision making algorithms. With the algorithms being capable of driving a car autonomously for a set real-world trials involving members of the public. Some of the most useful insight gained from this was not only how big the jump can be from theory and simulation to real-world implementation but also physical integration between hardware and bespoke software from a number of project partners with differing requirements.

Optimal Routing and Assignment for Commercial Formation Flight


Oct 2011 – Jul 2015 University of Bristol
PhD in the aerospace engineering department of the University of Bristol. Working under the headline of ‘future concepts in aviation’. Investigated potential fuel saving opportunities arising from commercial airliners flying in a close or extended formation; Looking to optimise routes on a global scale. By developing a fast analytic methodology for optimising rendezvous locations of formation members enables the macro/global problem of fleet assignment to become tractable.


Research students that I have supervised over the years.



Anas Shrinah

MSc Student


Jian Jiao

MSc Student


Karam Safarov

Phd Student


Xingyu Guo

MSc Student