BirdSense

The Device “BirdSense” is designed to prevents the birds from coming near your residence. especially in major cities.

BirdSense

The Device “BirdSense” is designed to prevents the birds from coming near your residence. especially in major cities.
sustainable cities and communities

SDG : 11 Sustainable Cities and Communities

Objective

The Device “BirdSense” is designed to put away the bird problem in your residence, especially in major cities. The device is equipped with a microcontroller, motion sensors, and an ultrasonic speaker. When installed on the balcony, It detects the movement of the incoming birds near the balcony and emits high-frequency sounds which helps in diverting the birds from coming towards your residence. These sounds are not audible to the human ears.

Solution

Version one will detect pigeons using a PIR sensor and use a speaker to produce  alarming or predatory sounds to scare the pigeon. 

Experiment 1.0  

To check whether the pigeon will be detected with the PIR sensor successfully, I setup an  ESP8266 in my balcony that logged the time on an Adafruit feed when the PIR sensor  returned a positive feedback. 

To verify whether the detection was only for a pigeon and not something else, a camera  was put to record the entire balcony. After the experiment I cross checked the time logs  and matched it to the video to see whether the detection was a false one or a true one. 

Unfortunately, the experiment was unsuccessful because multiple detections were logged  every minute. And almost none of them were pigeon detections. This happened despite  the PIR sensor being calibrated to minimum sensitivity and time delay. I suspect the  jumper being faulty. 

Experiment 2.0 

This time I used an Arduino for the sake of simplicity, changed all the wires and also used  a buzzer. 

This experiment was super successful and had zero false detections, the buzzer only buzzed when I put my hand in front of the sensor or a pigeon came. However, the only issue I learnt was that the PIR sensor would be able to detect a bird if it came from the range between the red lines drawn.

As drawn in the diagram above, if the pigeon enters far enough to the side that it doesn’t  cut the 3m long virtual cone space created by the PIR sensor, it will not be detected. This  blindspot hasn’t been a problem yet because pigeons usually enter through the centre.  But if they do figure out the blindspot, it will become a problem. To cover the blindspot,  two possible solutions are there: use two sensors and get a wider area of detection with  more accuracy as shown in the diagram above; or use different frenzel lenses. Option 2 is  definitely more cost effective and easier. But it’s doubtful how much of the blindspot will it  be able cover. 

Experiment 3.0 

Using a laser cutter, a small box was made from 3.5mm MDF wood. Inside the box was a  simple circuit with a Arduino Nano, buzzer and PIR sensor. The box was placed on the  pillar of a window facing outwards to detect for incoming pigeons. However, this  experiment wasn’t successful because it detected cars from the nearby street even after  being on the minimum sensitivity setting. 

 

Experiment 3.3 

Using the same prototype from the last experiment documented (E.3.0), the device was  placed on the balcony railing as shown below.

Since the sensor was still detecting the movement from nearby leaves on branches. I put  a small piece of plastic (the double sided tape’s removable covering) between the sensor  and the frenzel lens. This however reduced the accuracy of detecting an object in its  proximity but it also prevented any false positives. Since this particular area didn’t have  the pigeon problem, we just simply threw a beanie back and forth to simulate the pigeon  behaviour.

Project Contributors

Trishit

View detailed project documentation

This Project Has Been Developed By The Alumni Of Innovation School

Problem

Pigeons are a nuisance for people with balconies and open windows. They dirty the entire place with their feathers and droppings. Often the only solution for them is a pigeon net. Which however are very un-aesthetic. Nobody likes buying a gorgeous flat and putting a bird net on the windows in major cities like Mumbai.

Solution

Version one will detect pigeons using a PIR sensor and use a speaker to produce  alarming or predatory sounds to scare the pigeon. 

Experiment 1.0  

To check whether the pigeon will be detected with the PIR sensor successfully, I setup an  ESP8266 in my balcony that logged the time on an Adafruit feed when the PIR sensor  returned a positive feedback. 

To verify whether the detection was only for a pigeon and not something else, a camera  was put to record the entire balcony. After the experiment I cross checked the time logs  and matched it to the video to see whether the detection was a false one or a true one. 

Unfortunately, the experiment was unsuccessful because multiple detections were logged  every minute. And almost none of them were pigeon detections. This happened despite  the PIR sensor being calibrated to minimum sensitivity and time delay. I suspect the  jumper being faulty. 

Experiment 2.0 

This time I used an Arduino for the sake of simplicity, changed all the wires and also used  a buzzer. 

This experiment was super successful and had zero false detections, the buzzer only buzzed when I put my hand in front of the sensor or a pigeon came. However, the only issue I learnt was that the PIR sensor would be able to detect a bird if it came from the range between the red lines drawn.

As drawn in the diagram above, if the pigeon enters far enough to the side that it doesn’t  cut the 3m long virtual cone space created by the PIR sensor, it will not be detected. This  blindspot hasn’t been a problem yet because pigeons usually enter through the centre.  But if they do figure out the blindspot, it will become a problem. To cover the blindspot,  two possible solutions are there: use two sensors and get a wider area of detection with  more accuracy as shown in the diagram above; or use different frenzel lenses. Option 2 is  definitely more cost effective and easier. But it’s doubtful how much of the blindspot will it  be able cover. 

Experiment 3.0 

Using a laser cutter, a small box was made from 3.5mm MDF wood. Inside the box was a  simple circuit with a Arduino Nano, buzzer and PIR sensor. The box was placed on the  pillar of a window facing outwards to detect for incoming pigeons. However, this  experiment wasn’t successful because it detected cars from the nearby street even after  being on the minimum sensitivity setting. 

 

Experiment 3.3 

Using the same prototype from the last experiment documented (E.3.0), the device was  placed on the balcony railing as shown below.

Since the sensor was still detecting the movement from nearby leaves on branches. I put  a small piece of plastic (the double sided tape’s removable covering) between the sensor  and the frenzel lens. This however reduced the accuracy of detecting an object in its  proximity but it also prevented any false positives. Since this particular area didn’t have  the pigeon problem, we just simply threw a beanie back and forth to simulate the pigeon  behaviour.

Related Projects

A connected insole ( a wearable insole) for diabetic patients suffering from T2D consists of several pressure sensors to give dynamic feedback of the patients’ plantar pressure.
sdg 3

#sdginfocus

SDG 3 : Good Health and Well Being

sdg 3

#sdginfocus

SDG 3 : Good Health and Well Being

This project helps education institutions to track and improve the mental health of the educators.

#sdginfocus

#sdginfocus

Ever wondered what if an amputee with a prosthetic hand picks up a hot cup of coffee and drinks it? This feedback system let them know about the temperature of the surface.
sdg 10

#sdginfocus

SDG 10 : Reduced Inequalities

sdg 10

#sdginfocus

SDG 10 : Reduced Inequalities