Tricky Landing

The “Odysseus” robot lander created by Intuitive Machines seems to have landed successfully on the moon,  although reports say it is on its side rather than upright.  Details as of yet are quite sparse,  but depending upon whose reports you read,  it would appear the lander had a non-trivial and unintended horizontal speed at touchdown.  Odds are,  it was also tilted a bit in the direction of that horizontal motion.  It seems to have “tripped” on one of its legs being somehow obstructed,  overturning the lander as it touched down.

This is always going to be a serious problem for a robot moon lander,  as opposed to a manned craft.  It is still difficult-to-impossible to program a robot to do what a human pilot can do,  and robot vision is still nowhere near as good as human vision. 

Consider what Neil Armstrong had to do,  landing the Apollo 11 lunar module.  The computer-controlled trajectory was taking the vehicle into a tightly-packed field of multiple boulders as big as houses:  a guaranteed fatal crash!  Armstrong had to take manual control,  stop the descent into an unplanned hover,  and then direct that hover toward a clear landing site nearby.  The depletion of his rocket fuel was but a single handful of seconds away at engine shutdown.  

The rule-of-thumb stability criterion used for all successful landing leg-equipped craft on the moon and Mars,  is that the span between foot pads at least equals the height of the craft center of gravity,  and preferably exceeds it. The Odysseus lander only just barely met this,  and the also-recent Japanese lander did not meet it,  and was photographed upside-down after its landing!

It is not yet known what “tripped” the Odysseus lander,  but the odds favor either a leg striking a fixed rock or similar obstruction,  or else a landing pad digging into the surface dirt.  The “fix” for this is two-fold:  (1) increase the pad span to center-of-gravity height ratio significantly,  and (2) hinge the pads on the ends of the legs,  and spring-load them to tilt upward toward the radial-outward direction at each leg.  The first decreases the net overturning torque of a “tripping” incident,  and the second acts to prevent a pad from digging-in. 

Seems “intuitive” to me.  Maybe we old farts still have things to contribute,  after all. 

Update 2-26-2024:  The quote below,  from a PBS NewsHour story published this date,  confirms what I hypothesized about the Odysseus lander.  It apparently will cease operations tomorrow for lack of sunlight on its solar panels,  according to the story.  Highlighting is mine.

“The lander, named Odysseus, is the first U.S. spacecraft to land on the moon in more than 50 years, carrying experiments for NASA, the main sponsor. But it came in too fast last Thursday and the foot of one of its six legs caught on the surface, causing it to tumble over, according to company officials.”

What I found about the Japanese “SLIM” lander is enlightening,  although it is still unclear just exactly how it ended up on its nose.  It was supposed to hover and then tip over onto its side,  with its landing legs extending out that side.  In the long dimension,  pad span exceeds cg height,  meeting the criterion.  From side-to-side,  it does not meet the criterion,  a very real risk. 

But it did not fall over to one side,  somehow it instead went tumbling end-over-end,  which is the only way it could have ended up on its nose!  Some stories mention a problem with a main thruster (there were two in its bottom).  Those should be “off” during the actual tip-over-and-landing because they are too powerful,  so any main thruster-related hypothesis would have to have occurred before that process.

One possible main thruster-related hypothesis is that it may have experienced suddenly-asymmetric thrust just as it approached hover for final tip-over-and-landing.  If so,  that could have sent it tumbling end-over-end while still aloft,  instead of hovering into a controlled tip-over. There should be marks in the regolith if that hypothesis is true,  marks where it hit while already tumbling end-over-end.  In the low gravity,  it would have continued to bounce end-over-end after hitting the surface.  It just happened to quit bouncing,  while on its nose.  Improbable,  but possible.  Still,  only a hypothesis.