Please help with suggestions! I want to improve my prototype.

The goal; detect when a non-moving person is present from a distance of a few inches up to about 8 feet. The constraints are:

• No contact (no load cells)
• Battery powered - must run 6+ months on either (2) 9v batteries, or 4-6 AA batteries
• Can "sleep" for 1-2 minutes between checking for a person
• Detection range of a few inches to 8 feet
• Do not need to know the distance to the person, only need to know if they are present or not
• Must detect people wearing clothing, but sitting still
• Do NOT need to differentiate between a person and a large object that may appear
• OK to combine multiple sensor types (ultrasonic + IR, etc..)
• Target retail cost in low-volume production (few thousand units) under $50

Picture a chair or couch in a house. I need to detect, within a couple of minutes, when someone sits down, and when they leave. The sensor may be mounted on the chair/couch, or on a shelf or table near by. I don't care about false positives IF it is caused by a large object in place of a person (dog, large box, etc..).

My first prototype uses an ultrasonic sensor like this one. It meets all the criteria, except I would like better reliability at longer ranges. When someone is close to the sensor it works fine, but if someone is 6-8 feet away then their clothing (how much it absorbs the sound) and their body angle really affect the sensitivity. So I need either an additional, or an alternative sensor. All suggestions are welcome!

 

Have you given any thought to a PIR (passive infrared) sensor an electronic sensor that measures infrared (IR) light radiating from objects in its field of view. They are most often used in PIR-based motion detectors. They are popular with outdoor security lights and similar. They will detect a non moving person or animal when the object moves into their range. After that, however, unless the object moves, they will stop detecting until the object moves again. They will easily cover the range you mention and depending on the unit have a "field" of view. Here is an example of a basic inexpensive version you may want to check out the data sheet. I have seen that same unit on Ebay for about $3.00. From there they go up in cost depending on what you want as to range and field of view. The better the sensor the greater the cost and the higher the quiescent current draw so that needs considered. A Google of PIR Sensors should yield some results.

Ron

 

Thank you for the suggestion. I will check out the PIR sensors, but the limitation of not continually detecting a stationary person was the reason we passed over them initially. Maybe there's a way around that.

I thought of using an IR thermometer or IR camera but haven't found a sensor at a reasonable price yet.

 

Good luck in your quest. There should be something out there, likely IR, which will remain triggered for a stationary person or animal.

Ron

 

Micropower Impulse Radar can detect the motion of a person's chest as he or she breathes and more easily detect the presence of an object. These can be very inexpensive -cheap enough for the auto industry to use as backup warning detectors.

Here is the link to an brochure from the early days

This technology is also referred to as MIR.

 

Maybe there's a way around that.

A rotary/reciprocating shutter in front of the sensor?

 

A rotary/reciprocating shutter in front of the sensor?

That is a very good technique and is referred to as an optical chopper. One way to make one is to spin a metal disc between the sensor and the target, with a window cut out of the disc so that the sensor is alternatively exposed to the target and then the disc.

 

Thank you for the suggestion. I will check out the PIR sensors, but the limitation of not continually detecting a stationary person was the reason we passed over them initially. Maybe there's a way around that.

I thought of using an IR thermometer or IR camera but haven't found a sensor at a reasonable price yet.

You might look at a thermopile array, which is basically a low resolution thermal imager. Something like Panasonic's Grid-EYE (though it might be cost prohibitive for your application).

These work on the same principle as PIR, but by having multiple sensors in an array, you can detect temperature gradients. This, in turn, allows you to detect stationary objects or differentiate between multiple objects, people, etc.

 

Thank you everyone for the suggestions! I'll check these out and let you know what we end up with.

 

Thank you for the suggestion. I will check out the PIR sensors, but the limitation of not continually detecting a stationary person was the reason we passed over them initially. Maybe there's a way around that.

I thought of using an IR thermometer or IR camera but haven't found a sensor at a reasonable price yet.

PIR is good. You can get the distance but focus narrows. You didn't specify environment. PIR does not work so hood outside in summers where there is little difference between the person and the background. PIR sensors can detect changes, as mentioned, but can also detect static conditions with a change in the electronic.