Looking at what sensors to invest in for next season, I see two distance sensing modules that are of interest to our team. I know I don’t really want ultrasonic because it doesn’t seem to work past 10 feet and is prone to interference. The main thing I would use it for would be to calculate the distance to the goal for shooting. I saw a team earlier with a title along the lines of “very fast auto aim”(I can’t seem to find the thread…) and they were using LiDAR for their distance readings which made me think that LiDAR might be a cool investment.
So that leads me to my inquiry, is LiDAR worth the extra hundred dollars vs an IR sensor? The LiDAR seems to have more distance, but, for example, Stronghold, you really don’t shoot from farther than that. Thanks!
I realize you are almost certainly just repeating the names you saw the seller use, but … FYI, for future reference, remember that LIDARs often are IR sensors (sensors that use IR light)).
To answer your question properly, respondents are going to need more information about your target(s), your accuracy, precision and speed requirements, your plan for discriminating among multiple potential targets, the value of your time, the capabilities of your vehicle, the cost of of using machinery to augment the raw sensors, your skills and abilities, your wealth, etc.
With that said, I hope you can get useful information from folks familiar with each sensor.
Given that the total budget (over KoP) for an FRC robot is $4000 (R10 this year), it will rarely be the case that a sensor that costs over $100 will be cost effective. The one common example I can think of is an IMU (integrated motion unit) which really merges several sensors with a lot of processing.
For FRC range finding purposes, I have found relatively inexpensive IR sensors to be great for short ranges (up to 80 cm/30 inches) and ultrasonics for mid-to-long ranges (20 cm/8 in - 10m/30+ ft). If you really need something that includes both short and long ranges, often a combination of two relatively inexpensive sensors can be as effective as one higher-priced one.
In general, when I find that we are planning to spend more than about $25 on a single sensor, I make sure that we (and usually I) have spent several hours searching for a less expensive alternative that will meet the requirements. The great majority of the time, we find one.
Also consider cameras as range finders. If you are looking for a known-size target, can send two parallel laser beams, or can exploit some other geometrical tricks, you can use the size/separation as a range finder.
Edit:
++
Second Edit: for off-season experimenting, I suggest being a bit more frugal than you would be during build season. There’s more time and less risk, so try to learn to work with (and find the limits of) less expensive sensors before investing in the high-end ones.
We have done work with both of those sensors you linked to. HVA 3824 used the Lidar on their bot this year and had good experience with it. We used that long range IR sensor for part of the season, and it worked just like all the other Sharp sensors.
The Lidar will be good to ±1 inch. The IR sensor was not that accurate, but it varied by distance. You need a good calibration curve for the IR sensor which you can produce during the off-season. The Lidar provides the distance over I2C, so no conversion is done in your code.
I will provide a contrary view to some above. Experiment with the technology now while you have the time. Learn its limitations and also work past the problems of getting and processing the data. If you wait until build season to learn a new sensor, it can be too stressful. We buy new sensors in the summer and be sure we know how to use them so we can immediately deploy them if needed in build season.