A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Frank Reed
Date: 2016 Nov 1, 12:26 -0700
There's a basic principle of navigation, which I think we all know, and I have quoted and described a number of times recently, but it doesn't seem to have a name: "use all available navigational information". It's the best "positive" reason to use celestial. The Sun is there, so use it. Applying multiple sources of information to a navigation problem is the best way to avoid huge blunders. What should we call this? Something more expressive than simple redundancy?
While I'm thinking of it, this relates to a recent navigation error ...above the planet Mars. The ESA test lander, named Schiaparelli, plummeted to its doom on Mars two weeks ago. Though the details are unclear, it appears that the spacecraft's software made a small navigation error and determined that the lander was two meters above the martian surface when it was actually around two kilometers above the ground. The software promptly shut off the landing rockets, which had fired for only a few seconds instead of the expected thirty seconds, and the lander then enjoyed a nice freefall to the surface where it made a beautiful, smoky hole in the martian landscape. This was not the plan.
Oddly enough, this is the second time that an error in altitude navigation shut down the rockets of a Mars lander prematurely leading to its destruction. The Mars Polar Lander or MPL sent to Mars by NASA back in 1999 apparently had a similar fate. Like the ESA lander two weeks ago, it is believed that it shut down its rockets at altitude instead of at landing. In this case, the best model for the anomaly suggested that the deployment of the landing legs 40 meters above the ground caused a hard vibration which was interpreted by the sensor/computer/software system as contact with the ground. Like the Schiaparelli lander, MPL then plummeted to the surface. Dropping 40 meters is a little less impressive than dropping 2000 meters, but it's all the same in the end.
These were navigation errors. The spacecraft in both cases (as nearly as we understand it right now) believed that it was at or very close to zero altitude when in fact it was high above the ground. Landing on the thrust of rockets is an unforgiving system. Normally it's binary: the rockets are on or off. And once shut down, you better be darn sure you've got solid ground beneath your feet. When you're landing on rocket thrust, there is no graceful degradation. You can't "glide in" for a rough landing. If you shut down too early, you crash. And that qualifies as a "huge blunder". In both of these Mars landing attempts, one two weeks ago and the other seventeen years ago, the system apparently had only one sensor to determine that the vehicle was near the ground. Both of these landers could have been saved if they had somehow applied this basic principle of navigation: use all available navigational information. If there had been multiple methods of determining ground proximity, then these "huge blunders" could have been avoided.
Conanicut Island USA