Puchalski’s Laws of Field Robotics Debug & Deploy
Akin’s Laws of Spacecraft Design are a classic of the internet and engineering. After a recent re-posting of the Laws on Hacker News (and me recently finishing Asimov’s Foundation Trilogy,) I present: Puchalski’s Laws of Field Robotics — heavily cribbed from the Spacecraft Design Laws, because they’re fantastic. I’m sure I’ll update and expand upon these as I get more feedback from my friends in the community.
The Laws of Field Robotics
1. Engineering is done with numbers. An analysis without numbers is merely an opinion.
2. To design a robot right requires immense effort. This is why they should operate even when some parts malfunction.
3. Design is iterative. You always need one more iteration than you think you do.
4. Your proudest design solutions may not make the final cut. Learn to accept it.
5. Two points make a line. Three data points plot a curve.
6. All relationships appear linear if plotted log-log with a broad marker.
7. Lack of information is not an excuse for halting analysis.
8. If uncertain, make an educated guess. If it’s an emergency, just guess. But correct it when accurate data becomes available.
9. When a brand-new issue arises, it’s probably not a new issue, just a new description of a known issue.
10. Prior analysis is not gospel. Trust your methods and be original.
11. (Pacilio’s Law) The software doesn’t “work” until it works on the vehicle.
12. Past experiences offer reality checks. However, being too grounded might stifle innovation.
13. Assuming you’re the smartest in the field is folly. If your results seem too good to be true, they likely are.
14. Half of what you learn in theory might not apply in practice. Wisdom is distinguishing which half.
15. True understanding comes from repetition or teaching.
16. You never fully understand your system until the first time it’s deployed to a new operational environment.
