CluelessGenius wrote: ↑
Thu Dec 06, 2018 11:45 am
alright i think i have an idea.
but 1st can someone tell me if i can connect multiple ir sensors to one pi and preferrably differentiate between them?
heres what i thought might work. what do you guys think? i am still not sure about increasing the resolution / accuracy and how to do it best but in generell i think this is the best option so far because the components needed are small enough that i could hide the pretty well
I think such a scheme would work much better with an array of LIDARs or TOF sensors. A scan of the items placed on the shelf before any people are around can be made. When people are around, a-after-person-has-moved-on-scan can be made to determine if the object is still in position. Each LIDAR in the array could be addressed through a MUX, such as a CD74HC4067 16-Channel Analog Digital Multiplexer module. The LIDARs, such as a TFMini, could be placed up to 400cm away from the objects to be monitored, for an object depth of scan of 100cm. The longer the LIDAR distance the higher the price. A TFMini has a beam width of 2 degrees, the more narrow the beam width the higher the price.
Once initial scans are made only periodic scans need be made. Such as adding in a few Micro Wave Motion detectors, you'd only need to scan the motion detected areas, after motion stoppage, to see if your objects have been moved. A motion detector could, also, trigger a video device to begin recording.
I, also, think an optical flow sensor array, could work as well, in the same sense of a LIDAR array.
I use a TFMini to scan an area, with a rotating servo, to produce images like the motion detector used in that Aliens movie.
Without knowing why you are deleting my postings, I will not know how...