Well the title says it all, and I am here out of desperation! I have tried every commercial product (ultrasonic, water sprayer, pellets, smells, plants etc) to no avail. So as an electronics and software design engineer of 30 years I have decided to attempt to build my own system. Whilst I appreciate it will require a fair amount of effort and probably cost if it means my family can use our garden without getting cat mess on them it will be worth it.
So I am here really for any suggestions about how best to achieve this. From my own experience I have discovered that water based systems have been the most effective. Therefore, I think a system that sprays water is likely to be the most successful.
The issues with the commercial PIR based water systems are multiple false triggers, and failure to trigger at all if an object moves into sensor area slowly enough. But when it does work, it works well. So I think I need to improve on this type of concept.
In order to get maximum coverage of my garden I plan to use a series of water sprays, all linked together and controlled by an MCU. I will archive this by using irrigation hose buried in the ground Tee’d off to each ‘nozzle’. The nozzle will have a solenoid valve to control water on / off. That’s the easy bit.
In my day job we use thousands of PIC processors (8 to 32 bit) so will use one of those to control everything – probably an 8 bit, not sure it will need the might of a 32 but I see what is going spare!
Now the difficult bit, I need to be able to reliably sense the presence of a cat (or fox) when it moves into a hot zone. I do not want to use a PIR as these do not seem reliable. In an ideal world I would have an IR emitter one side of garden and IR receiver the other and when broken the relevant solenoid is triggered. But I am not sure this would work over a 15 to 20 foot range and even if it did it would need careful physical alignment. Is there a sensor I could use that is single sided (a bit like an IR reflective sensor)? May be some kind of IR heat sensor? I know Omron do some but they are quite expensive (so yes cost is an issue I guess) I would need about 10 to cover my garden.
My other thought is to mount a USB IR surveillance camera on the wall of the house such that it has a good view of the garden. Then connect this to a PC. Then I could process the image looking for changes across a series of frames, mapped against the various zones. Then send message to my PIC to fire relevant sensors. This seems to be a good solution as I could visually see things whilst developing the code and tweak accordingly. But would require a PC running 24/7 – may be a Raspberry PI could do this.
I am aware I sound quite mad, but this is really a very frustrating and disgusting problem.
I would appreciate any input or thoughts about how I could humanely solve this issue – I can’t be alone!
My background is in microelectronic design of medical equipment and computer software. I have access to full PCB prototyping and assembly facilities as well as hardware prototyping and machining.
Thank you!
So I am here really for any suggestions about how best to achieve this. From my own experience I have discovered that water based systems have been the most effective. Therefore, I think a system that sprays water is likely to be the most successful.
The issues with the commercial PIR based water systems are multiple false triggers, and failure to trigger at all if an object moves into sensor area slowly enough. But when it does work, it works well. So I think I need to improve on this type of concept.
In order to get maximum coverage of my garden I plan to use a series of water sprays, all linked together and controlled by an MCU. I will archive this by using irrigation hose buried in the ground Tee’d off to each ‘nozzle’. The nozzle will have a solenoid valve to control water on / off. That’s the easy bit.
In my day job we use thousands of PIC processors (8 to 32 bit) so will use one of those to control everything – probably an 8 bit, not sure it will need the might of a 32 but I see what is going spare!
Now the difficult bit, I need to be able to reliably sense the presence of a cat (or fox) when it moves into a hot zone. I do not want to use a PIR as these do not seem reliable. In an ideal world I would have an IR emitter one side of garden and IR receiver the other and when broken the relevant solenoid is triggered. But I am not sure this would work over a 15 to 20 foot range and even if it did it would need careful physical alignment. Is there a sensor I could use that is single sided (a bit like an IR reflective sensor)? May be some kind of IR heat sensor? I know Omron do some but they are quite expensive (so yes cost is an issue I guess) I would need about 10 to cover my garden.
My other thought is to mount a USB IR surveillance camera on the wall of the house such that it has a good view of the garden. Then connect this to a PC. Then I could process the image looking for changes across a series of frames, mapped against the various zones. Then send message to my PIC to fire relevant sensors. This seems to be a good solution as I could visually see things whilst developing the code and tweak accordingly. But would require a PC running 24/7 – may be a Raspberry PI could do this.
I am aware I sound quite mad, but this is really a very frustrating and disgusting problem.
I would appreciate any input or thoughts about how I could humanely solve this issue – I can’t be alone!
My background is in microelectronic design of medical equipment and computer software. I have access to full PCB prototyping and assembly facilities as well as hardware prototyping and machining.
Thank you!