COLREGS Autonomy

Funded by the Office of Naval Research, ONR 311, Dr. Behzad Kamgar-Parsi. ONR 333, Dr. Robert Brizzolara (through Spring 2018)

The collaboration that governs the interactions between any two manned vessels has evolved over the centuries into a well-established protocol and codified regulations that establish roles and responsibilities. To date, no protocol or similar regulation has been established for unmanned vessels, or more specifically, for autonomous vessels that are not under command of a human operator. As the expansion of unmanned surface vehicles (USVs) into the manned vessel domain continues, a well-established integration plan will need to be developed, codified, tested, and promulgated to the mariners and autonomous designers around the world. With the recent development of manned vessel rule adaptations to the autonomous marine vehicle decision space, a push is being made to incorporate unmanned vessels into the already established realm of manned collision avoidance rules and regulations known as COLREGS. Before this transition can happen, advancements in the ability of USVs to comply and show compliance with COLREGS are required.

This project develops (1) methods to further the ability to autonomously control a single USV while balancing multiple mission objectives, including COLREGS alongside mission-specific goals of searching, patrolling or intercepting a contact. (2) It will extend these capabilities to multi-USV scenarios with USVs acting in coordination. (3) It will establish a form of COLREGS compliance metrics that includes both the measure of collisions or near-collisions alongside the measure of mission efficiency. (4) It will also explore edge-case situations where the decision not to follow the COLREGS or bend the COLREGS may be necessary in cases of extreme mission urgency.

Principal Investigators:

  • Michael Benjamin (PI)
  • Henrik Schmidt (Co-PI)
  • John Leonard


  • Kyle Woerner
  • Joe Leavitt

Publications COLREGS

xxxi2018 (2 items)

  1. Michael R. Benjamin, Capturing Velocity Function Plateaus for Efficient Marine Vehicle Collision Avoidance Calculations, OCEANS 2018 MTS/IEEE Kobe Japan, May, 2018. (bibtex)
  2. Kyle L. Woerner, Michael R. Benjamin, Real-time Automated Evaluation of COLREGS-Constrained Interactions Between Autonomous Surface Vessels and Human Operated Vessels in Collaborative Human-Machine Partnering Missions, OCEANS 2018 MTS/IEEE Kobe Japan, May, 2018. (bibtex)

2017 (4 items)

  1. Michael R. Benjamin, {Fast-CPA}: A Layered Caching Algorithm for Rapid Closest Point of Approach Calculations in Marine Collision Avoidance, OCEANS 2017 MTS/IEEE Anchorage, Anchorage, AK, September, 2017. (bibtex)
  2. Michael R. Benjamin, Autonomous COLREGS Modes and Velocity Functions, MIT Computer Science and Artificial Intelligence Lab, MIT-CSAIL-TR-2017-009, May, 2017. (bibtex)
  3. Kyle L. Woerner, Michael Novitzky, Michael R. Benjamin, John J. Leonard, Legibility and Predictability of Protocol-Constrained Motion: Evaluating Human-Robot Ship Interactions Under {COLREGS} Collision Avoidance Requirements, In Workshop on Mathematical Models, Algorithms, and Human-Robot Interaction at {RSS 2017}, Cambridge, MA, July, 2017. (bibtex)
  4. Joseph W. Leavitt, Intent-Aware Collision Avoidance for Autonomous Marine Vehicles, Massachusetts Institute of Technology, June, 2017. (bibtex)

2016 (2 items)

  1. Kyle L. Woerner, Michael R. Benjamin, Michael Novitzky, John J. Leonard, Collision Avoidance Road Test for COLREGS-constrained Autonomous Vehicles, OCEANS 2016 MTS/IEEE Monterey, 1-6, September, 2016. (bibtex)
  2. Kyle L. Woerner, Multi-Contact Protocol-Constrained Collision Avoidance for Autonomous Marine Vehicles, Massachusetts Institute of Technology, 2016. (bibtex)

2015 (1 items)

  1. Kyle L. Woerner, Michael R. Benjamin, Autonomous Collision Avoidance Tradespace Analysis for High-Speed Vessels, 13th International Conference on Fast Sea Transportation. Society of Naval Architects and Marine Engineers, 2015. (bibtex)

2014 (2 items)

  1. Kyle Woerner, COLREGS-Compliant Autonomous Collision Avoidance Using Multi-Objective Optimization with Interval Programming, Massachusetts Institute of Technology, Cambridge MA, June, 2014. (bibtex)
  2. Kyle Woerner, Michael Benjamin, Safety and Efficiency Analysis of Autonomous Collision Avoidance Using Multi-objective Optimization with Interval Programming, Naval Engineers Journal, (126)4, 163--168, 2014. (bibtex)

Page built from LaTeX source using texwiki, developed at MIT. Errata to Get PDF