7
$\begingroup$

Some spacecraft designed for autonomous landing / hazard avoidance (such as Surveyor) use three-beam Doppler radar for range-rate measurements, and some (such as Morpheus) have utilized a three-beam Doppler lidar for the same.

Are there advantages and disadvantages to each, such as sensitivity, power, range, etc.?

Note: There has been some confusion among answerers about the type of sensor I'm referring to. A Doppler LIDAR (example) is not the same as a LIDAR. Whereas a LIDAR typically has a single beam that scans or simultaneously shoots lasers at many points (returning a point cloud), a Doppler LIDAR usually has only a few beams which are aimed in fixed directions. A LIDAR returns only the range to each strike-point. A Doppler LIDAR returns the range as well as the range-rate of the strike-point. I believe the same is true for Doppler radar.

$\endgroup$
  • 1
    $\begingroup$ @uhoh Thanks for the feedback. I just added some links. $\endgroup$ – Doctor Mohawk Oct 14 at 20:54
  • $\begingroup$ looks good, thanks! $\endgroup$ – uhoh Oct 15 at 2:36
  • $\begingroup$ It is the inherent properties of the photons that creates the differences, not the design of the sensor (the number/timing/direction of the photons emitted and the signal processing). A Doppler LIDAR and a doppler radar will have similar relative benefits/cost as any other sensor design as photon emission and processing can be adjusted similarly for either. $\endgroup$ – johnDanger Oct 17 at 22:45
  • 1
    $\begingroup$ @uhoh Thanks for doing the bounty! $\endgroup$ – Doctor Mohawk Oct 18 at 17:09
  • $\begingroup$ The only difference between lidar and doppler lidar is the doppler part... Just like the only difference between radar and doppler radar is the doppler part. All that doppler means is, it's measuring change in rate of return along with everything else. When comparing the differences between doppler lidar and doppler radar, the only differences are in the lidar and radar parts, not in the doppler parts... That is, I'm curious why this needs more attention, given johnDanger's answer covering the differences between lidar and radar... $\endgroup$ – Ghedipunk Oct 18 at 18:45
7
+50
$\begingroup$

NOTE: This discusses autonomous car sensors but the same physics will apply to any longwave vs. shortwave sensor comparison (cost being the wildcard)

An interesting parallel is Tesla Inc.'s decision to use radar as the active component for their autonomous driving system rather than (much more common) LIDAR. My recollection of the costs/benefits of radar over LIDAR given by Tesla Inc. CEO Elon Musk in several interviews/speeches are

Radar Benefits:

  1. cost, LIDAR systems can cost 10 - 100+ times more than radar systems

  2. penetration of certain materials, radar can better "see" through rain, dust, clouds. Or as Mr. Musk puts it

Photons of [radar's] wavelength travel easily through fog, dust, rain and snow, but anything metallic looks like a mirror.

  1. range, following #2, radar (radio waves) travel a much further distance in an atmosphere without absorption

  2. power consumption, each radar signal requires less power to send due to its lower frequency (a photon with a wavelength in the visible spectrum has about a million times more energy (~1eV) than one with a wavelength in the radio spectrum (~10E-6 eV). This energy had to be expended by the instrument to create each photon)

LIDAR Benefits:

  1. increased resolution, LIDAR can map objects much more accurately than radar due to the difference in wavelength. Mr. Musk describes this problem with radar

A discarded soda can on the road, with its concave bottom facing towards you can appear to be a large and dangerous obstacle


As for spacecraft landing systems, they may be operating in varying environments and need the system that suits their specific situation.

  • Is high spacial resolution required to avoid rocks or slopes? Pick LIDAR

  • Is there a power or cost constraint? Pick radar

  • Is there dust or liquids in the atmosphere that must be penetrated? Pick radar

  • Does the system need to operate at a high altitude in an atmosphere? Pick radar

I would imagine an ideal system would use radar for the high altitude portions of a descent and LIDAR for final landing maneuvers.


The quotes from Mr. Musk were taken from his blog post at https://www.tesla.com/blog/upgrading-autopilot-seeing-world-radar

Below are two articles discussing the described differences in further:

  1. https://medium.com/@intellias/the-ultimate-sensor-battle-lidar-vs-radar-2ee0fb9de5da

  2. http://robotsforroboticists.com/lidar-vs-radar/

Especially interesting is the following picture from article 2 showing the difference in resolution between LIDAR (whose frequency is near the visible spectrum) and radar.

enter image description here

$\endgroup$
  • $\begingroup$ Additionally, Lidar is typically more mechanically complex and has moving or spinning parts. Radar can be constructed to be completely static. Moving parts are always a failure point and avoiding them is usually a good engineering decision $\endgroup$ – Dragongeek Oct 17 at 8:02
  • $\begingroup$ @Dragongeek this is true for autonomous road vehicles but I would think that any spacecraft would have a fixed, downward facing LIDAR assembly. This would of course narrow the cost difference between the two systems. $\endgroup$ – johnDanger Oct 17 at 15:18
  • $\begingroup$ Excellent links, thanks! $\endgroup$ – uhoh Oct 17 at 15:45
  • $\begingroup$ Hey, this is good information, but not totally applicable. Doppler LIDAR is a different sensor than a LIDAR sensor — usually it's got one to three beams, no moving parts, and measures range-rate to the target rather than just range. I believe the same is true for Doppler radar. $\endgroup$ – Doctor Mohawk Oct 17 at 20:19
  • 1
    $\begingroup$ The general physics of shortwave vs. longwave active sensors holds true for any form factor (cost being the one exception, hence the wide range given). The difference between various sensors of the same wavelength (Doppler, rangefinder, 3-D imager) is going to be in number/direction of photons emmited and the data processing algorithms used. The behavior of the photons themselves will be unchanged. $\endgroup$ – johnDanger Oct 17 at 22:32
2
$\begingroup$

I think you mean to ask the question about LIDAR as used for velocity measurement in spacecraft rather than the LIDAR used in autonomous vehicle. Here, I list some few points relating to both altimeters and velocimeters both.

Ka-Band Radars/Velocimetry

  1. It has advantage that, one can actually get sort of averaged information about altitude and velocity over the rough surface due to beam width

  2. Over higher altitudes, these sensors have slightly degraded performance mainly due to signal strength

  3. Power consumption is less, therefore allowing for higher frequency data

  4. At lower altitudes, it has problems of multipath

  5. Though at lower altitudes it can actually penetrate the dust(rising as you come close to surface)

  6. Can penetrate engine plume therefore never to worry about engine plume cone over full throttling regime

  7. The terrain effects are problem for the radar doppler

Laser altimeter/Doppler Lidar

  1. Can provide pin point altitude, therefore any compensation done for local terrain is possible

  2. The accuracy almost remains same for full operating range

  3. Can provide data till very low altitudes

  4. High power consumption. Time consuming as laser needs to be pumped for each operation. Higher frequency measurements bear risk for damaging the sensor

  5. Plume/Dust reflection is a real worry !

  6. The terrain variation is of no concern here. It gives true line of sight velocity !

Note: I am unable to find source for information as these points are by experience than actual source. So take it with grain of salt.

$\endgroup$
  • 1
    $\begingroup$ I've added some clarifying information to the question. I'm not sure you all are talking about the same class of sensor as I. $\endgroup$ – Doctor Mohawk Oct 17 at 20:24
  • $\begingroup$ Additionally, an optical device requires some sort of aperture. Lens ports could become obscured by dust, scratched, or damaged. A radar module can be built inside of a solid component and rangefind through an exterior cover which is made of something like plastic $\endgroup$ – Dragongeek Oct 22 at 13:19
  • $\begingroup$ @dragongeek I do not think dust is issue much for space missions, as it is just one time landing by the time you come close enough, generally below 10 m, only inertial nav is used till touchdown, there is not much scope to get dust. Unless, we have dragonfly type of mission, then it maybe a worry I think $\endgroup$ – Prakhar Oct 22 at 14:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.