Some notes about Optical Distance sensors
Page 1 of 1

Author:  dtrotzjr [ Mon Jun 25, 2007 12:09 am ]
Post subject:  Some notes about Optical Distance sensors

I spent some time this weekend writing a driver for the Mindsensor Dist-NxL sensor. One of the things that caught my attention is the way that this sensor works.

The sensor works by emitting a light from one 'eye' and catching it on the other. The intensity returned is converted into a voltage, the greater the voltage, the smaller the distance is returned. This makes sense intuitively when we think about it. The farther we are from an object the darker the light intensity will be that bounces from that object.

The problem with this type of sensor is that while it will give you distances in terms of millimeters the error varies greatly depending on what you are measuring. For example a dark object will absorb more of the light and give bigger distances than objects that are white. In fact lighting in a room can also play a role in the values you get back so you have to treat these values with some care.

The only hope is if you have a very controlled environment, i.e white walls, consistent lighting, and you recalibrate the sensor for this environment, but this is not realistic, instead we will need to weigh the values we get against our knowledge of the situation and use them accordingly.

Any comments?
David Trotz

Author:  starwarslegokid [ Mon Jun 25, 2007 2:36 pm ]
Post subject: 

Does the lighting conditions really affect the sensor? I thought the color of the objects would be the biggest problem because they will absorb and reflect IR wavelengths differently. Visible light should be in wavelengths not detectable by the receiver (I would hope) and shouldn't have any affect on the sensor. The only exception I see would be sunlight because it gives off both visible and infrared light.

That would be a good experiment, try high light, low light, and sunlight conditions on a single colored object of a fixed size and compare there values. Have you tried anything like this yet and have any data? I did some experiments a year ago on the Techno-Stuff Dual IR Sensor and got that different colored objects could yield up to a two inch difference in detection.

Thanks for the observations B-)

Page 1 of 1 All times are UTC - 5 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group