Distance sensor in movement

[Kailegh]

Hi all!! I am trying to make a king of autonomous car (just 15km/h). I need to measure its distance to a wall which will always be 20-30 cm, I need to measure with accuaracy of at least 1mm. I have looking different kind of lasers, but it accuaracy it is at the limit, and I am no very sure if they can work when moving at 15 km/h. I do not know if it would be better to use ultrasounds.

I came looking for inspiration!

Thank you!!

 

[Gryd3]

Hi all!! I am trying to make a king of autonomous car (just 15km/h). I need to measure its distance to a wall which will always be 20-30 cm, I need to measure with accuaracy of at least 1mm. I have looking different kind of lasers, but it accuaracy it is at the limit, and I am no very sure if they can work when moving at 15 km/h. I do not know if it would be better to use ultrasounds.

I came looking for inspiration!

Thank you!!

I will first voice my concern of course, that experiments with autonomous cars should be left to experienced knowledgeable people... but if you are working on a model car, then I would encourage you to look at other types of laser... and am curious why you need need to have that high of accuracy when you are measuring values 200-300 times higher than that.
Laser, Sonar, etc. will all have trade-offs... the further you go, the less accurate you will be,
So. consider perhaps what would the problem be if you measure 20.2 cm as opposed to 20.0 cm? Would getting this type of reading break anything?

 

[Amar Dhore]

To measure the distance in moving cars, ultrasonic sensors would be my first option. Most of the car companies uses these sensor for Ant-collision as well as other cars/object detection.

 

[Kailegh]

I will first voice my concern of course, that experiments with autonomous cars should be left to experienced knowledgeable people... but if you are working on a model car, then I would encourage you to look at other types of laser... and am curious why you need need to have that high of accuracy when you are measuring values 200-300 times higher than that.
Laser, Sonar, etc. will all have trade-offs... the further you go, the less accurate you will be,
So. consider perhaps what would the problem be if you measure 20.2 cm as opposed to 20.0 cm? Would getting this type of reading break anything?

it is a project, not a real car, just about 25 kg, it needs to measure a distance (that is its primal target) to be able to recognize damages in an enviroment ( those are the requirements, not of my choice) from about 20 cm with a resolution less than 1mm. I have been told that errors bigger than 1mm are not acceptable. I have used lasers to measure from fixed positions, but my doubt would be if they can also work on movement, for example:

https://www.sick.com/gb/en/distance...ment-sensors/od-value/od2-p120w60c0/p/p185890

I have heard that ultrasonics have worse resolution than laser, but I will give them a look.

Can I use these sensor in not normal direction to the surface I want to measure? I mean, for example with 60° incidence?

PS I forgot to say that I have to measure the distance to the lateral of the "car", neither in front nor behind.

Thank you all for your answers

 

[davenn]

To measure the distance in moving cars, ultrasonic sensors would be my first option. Most of the car companies uses these sensor for Ant-collision as well as other cars/object detection.

only for close up stuff .. to stop reversing into walls, car parked behind you etc

full speed anti-collision systems use microwave RF systems at 76 GHz

 

[davenn]

from about 20 cm with a resolution less than 1mm. I have been told that errors bigger than 1mm are not acceptable

and I seriously doubt you are going to be able to achieve that accuracy

I service survey equipment that has accuracies distance measurement down to 1mm and angular measurement down to 1 sec. These units cost $40,000 +

If your small error is going to be achievable ... it isn't going to be something basic found on any forum
its going to be a very expensive hi-tech system designed by highly experienced electronics engineers

Dave

 

[Gryd3]

and I seriously doubt you are going to be able to achieve that accuracy

I service survey equipment that has accuracies distance measurement down to 1mm and angular measurement down to 1 sec. These units cost $40,000 +

If your small error is going to be achievable ... it isn't going to be something basic found on any forum
its going to be a very expensive hi-tech system designed by highly experienced electronics engineers

Dave

Well... I'm not very familiar with it, but what about Lidar, or a hybrid... using multiple sensors and combining their results to form a more accurate result?