Athabasca University | AU Student/Staff Login | Invited Guest Login
POST - Autonomous Surface Vessels.
POST - Autonomous Surface Vessels.
In their paper, Males et al. (2022) survey the use of MAS in surface vessels. This is a goal that has been considered for a rather long time (approximately 130 years with Tesla). The main reasons for automating vessels are to cut costs and to increase safety (since most accidents are caused by human error). In some ways, the development of automated land vehicles has similar MAS issues, but not so in other ways.
The survey looks at papers in the following summarized categories:
1) Navigation – the combination of technology, map data sets, and regulations make this a natural area to benefit from MAS
2) Safety – a number of safety issues are considered such as avoiding collisions, preventing fires, thwarting piracy, handling mechanical failures (engine, hull, etc), and avoiding and recovering unanticipated grounding.
3) Overall monitoring – this involves a special agent that with the objective of monitoring individual vessel behaviour to ensure regulatory compliance.
In general, development of autonomous surface vessels is in early stages, with many of the above functions not well studied.
Males, L., Sumic, D., Rosic, M. (2022) Applications of Multi-Agent Systems in Unmanned Surface Vessels. Electronics 11, 3182. https://doi.org/10.3390/electronics11193182
Replies
- Gaganpreet Jhajj December 6, 2023 - 1:24am
Hi Duncan,
I found your discussion on MAS in autonomous surface vessels quite interesting. One thing I immediately thought of in this context was drone light shows, which are becoming increasingly commonplace, incorporating ideas from MAS, swarm robotics, and other related fields. Drone shows are becoming much more common as a somewhat greener alternative to the fireworks we often use during the holiday season [1]. It is becoming increasingly widespread; for example, a recent Esports event in Korea had a drone show that lasted over ten minutes and was viewed by millions worldwide [2,3].
Autonomous drones, such as those in light shows, are often relatively small and take place far away from populations, so the overall risk of a crash and other complications are often controlled. However, there is continued research on optimizing such formations, and recent work has explored new algorithms for allocating drones to target destinations better [4].
Additionally, similar to the lead follow characteristics present in flocks of birds, research has been done into creating similar lead follow systems [5], which can decrease both resource and computational requirements in complex drone arrays. Various companies and militaries are also investigating lead-follow systems such as this to have land-based autonomous units that can follow troops.
The progress in aerial vehicles via MAS could proceed much faster. In the sphere of entertainment and military applications, we see more and more usage of automated systems. In civilian applications, more sophisticated autopilot features are used. I do agree that surface and land-based vehicles are still a while away before we see true systems available for self-driving, for example.
Sources:
[1]N. W. · C. News ·, “Drone light shows a potential firework alternative during wildfire season | CBC News,” CBC. Accessed: Dec. 06, 2023. [Online]. Available: https://www.cbc.ca/news/canada/ottawa/drone-light-show-fireworks-canada-day-1.6886497
[2] “Drones light up the night sky, delivering holiday cheer in December - CBS Texas.” Accessed: Dec. 06, 2023. [Online]. Available: https://www.cbsnews.com/texas/news/drones-light-up-the-night-sky-delivering-holiday-cheer-in-december/
[3] Drone light show at Hangang Park in Seoul | League of Legends Champions Korea 2023. Accessed: Dec. 06, 2023. [Online Video]. Available: https://www.youtube.com/watch?v=o11Pz9_aYXQ
[4] D. Nar and R. Kotecha, “Optimal waypoint assignment for designing drone light show formations,” Results in Control and Optimization, vol. 9, p. 100174, Dec. 2022, doi: 10.1016/j.rico.2022.100174.
[5] A. Lanteigne, E. Kibru, S. Azam, and S. Al Shammary, “Design of a Drone Lead-Follow Control System,” in 2017 Systems and Information Engineering Design Symposium (SIEDS), Charlottesville, VA, USA: IEEE, Apr. 2017, pp. 162–167. doi: 10.1109/SIEDS.2017.7937709.
COMP667 Multiagent Systems
