Monthly Progress Blog: January
Drew Charter School Lemelson-MIT InvenTeam
300 E Lake Blvd
SE, Atlanta, GA 30317
January 31, 2017
Project Description Update
This month we have broken into groups, and are working on separate tasks to complete and eventually come back together. All groups had something that needed to be researched, some parts needed to be ordered and some sensors needed to be tested. The testing teams have been working tirelessly to get their sensors functioning in mock conditions. Our goal is to complete as much as possible before the Mid Grant Tech Review (MGTR), so when we present to our visitors we will receive the most efficient and informational feedback.
Our project description remains the same, as we are still determining the feasibility of our proposed design. Our project description is:
Our device will mitigate the number of hot car deaths of children and pets. We will investigate and test a variety of technologies to integrate into a simple, reliable device with a three tiered alert system. First, a text alarm will be sent, next, a voice alarm for individuals in the vicinity will be activated, and finally a GPS location of the vehicle will be sent to authorities notifying them of the emergency.
This month, we conducted an in depth beneficiary survey to generate a more broad understanding of the needs and desires of perspective product users. Overwhelmingly, the responses to our survey supported our three tiered system.
Progress During This Month
In each of our sensor teams, we have continue where we left off from December. In order to make the greatest progress, we have had to pick up the pace in everything that we do in preparation for the MGTR.
-Infrared Sensing Technology
Naomi Ray is the one taking charge on getting the infrared sensor up and running. She has faced many setbacks related to finding the correct connector to interface with the Arduino. Early this month, the correct connector came in and the sensor was quickly up and running. Naomi quickly learned that some additional coding would be required. Our sensor creates a list of temperatures that it sees within its viewframe. It does not interpret the data that it collects. From there, Naomi has worked with our software and arduino expert, Mr. Ayers, to generate some logic to get the infrared sensor to interpret the 16 data points it reads every second to determine if a human is present. The code logic relies on there being a difference in the 16 temperature values from the infrared sensor. The sensor detects 16 temperatures in a 4 x 4 grid. So the code will scan the lowest and highest temperature values and then take the difference between them. If the difference is higher than a certain number of degrees, then the serial monitor will print “ Human Presence Detected" and if the difference is lower than the predetermined value, then the serial monitor will print “ No Human Presence Detected.”
This has been an iterative process, but on Friday January 27th, Naomi was able to begin testing the sensor in the car. It was able to sense people seated in the middle and side seats in the back. Testing will continue next class period. Congratulations Naomi! Your persistence with this task has been impressive! It appears that the IR sensor is going to be a viable option!
-Motion Sensing Technology
Our motion sensing team has been spearheaded by Jordan Wilson. This month, Jordan has tested his motion sensor in the car. The initial testing procedure included having one student sit in various seats in the car and move their arm. Results were recorded for each position in the vehicle. Since it is our goal to protect children and pets, we tested in the back seat and the front seat of Ms. White’s Sedan. Another test we conducted was aimed at determining if a pet would be sensed if it was on the floorboards of the vehicle. The motion detector sensed all motion except moving feet on the floor board in the front seats.
One other important additional test that was conducted was to rule out the possibility of a false alarm caused by the motion of someone outside of the vehicle. The motion sensor was tested to see if it would detect a passerby, and it the tests conducted, it did not.
It was then suggested by Mr. Ayers, that we refine our test to more closely mimic the motion of a baby or a pet. To do this, we needed something smaller, with smaller movement than a teenager. We ordered a remote control dog. We conducted the testing again to verify that something significantly smaller would be detected by our motion sensor. Unfortunately, the motion detector was not sensitive enough to detect the motion of the remote controlled dog.
In our three tiered alert system, we will need an integrated GPS so that we can notify the authorities of the GPS location of the vehicle in the case of an emergency that has received no responses to the other two tiers.
During the month of January, after many times of trying to receive a location from the GPS module we were unsuccessful, we had to wait for a long time for parts to come in because we ordered them from China. We got the code and we were able to make the GPS code communicate with the arduino. Obstacles that we faced includes things such as improper soldering, error in coding, and receiving parts late.
-Temperature Sensing Technology
The temperature team led by Jeffrey Barber has a functioning, efficient sensor that is ready to be incorporated with the other aspects of the product. Throughout the month of December before we left for winter break, the temperature team made sure that the sensor was working correctly and accurately. In addition, we wanted to make sure that we could successfully sense the temperature and at critical points that the sensor would trigger an alarm. The team used arduino software to develop the coding needed for our sensor to function properly. Also we used wires, plug-ins, resistors, and a breadboard so the temperature sensor along with the buzzer would work simultaneously so that it would be as efficient as possible. The obstacles or errors we faced includes creating the correct arduino code so that their sensor would work the way it was needed. Another obstacle includes wiring and placing each piece such as the sensor, buzzers, and plug-ins in their specific positions so the sensor will be aligned with the buzzer and would respond when the temperature reaches a certain point.
-Additional research completed this month
How does the sensor know when to sense?
This month, we began researching how exactly we will get the sensor to work when the vehicle occupants have left the vehicle, but not when there is an adult in the vehicle. Additionally, we know that we don’t want to receive notifications when the vehicle is on. We brainstormed ways our device can gather this information, and are currently leaning towards a build that integrates with existing vehicle electrical control units. To simulate this, a new technical team has emerged. This team is led by Blake Turpeau, and has the mission of creating a mock “ignition” circuit and a mock “weight sensor” circuit. we have been working closely with Mr. Ayers to get their circuit functioning with a readout that indicates that it is appropriate to enable the sensors.
After talking with a veterinarian over Christmas break, it was suggested that our team look into Capnograph technology as an alternative to motion or IR to indicate the presence of a child or pet. Capnography is typically used in the medical field to assess CO2 output of patients. Xavier researched Capnograph technology to determine if it is a viable technology to indicate the presence of a human or pet in a vehicle. Through research, Xavier converted potential PPM of CO2 generated by an adult over the course of a few minutes in a car. The results demonstrate a potential to use this technology as a third resort if needed.
Electronic Control Units:
This month, the question has arisen that we may need to plan to integrate our technology into new vehicles instead of offering it as an after market product. Reasons for this include the need to know when the device should be on or off. We will continue to research this as an option, and have reached out to several car manufacturers in the area for technical support.
Jurnee Miller has now taken the lead on the task of getting the device to send a text to our phone when “presence” is indicated. She has researched materials needed, and it appears that the best route is going to be to use an arduino with Wifi. These arduinos have been ordered and the legwork to get them integrated has been done. We are hoping to receive an SMS text indicating “presence” of a child or pet in the vehicle soon.
The second tier of our alert is to send an audible alarm to people in the vicinity that the occupant in the vehicle is in danger. Jeffrey Barber has taken the lead on learning how to accomplish this. He has learned that we need an “arduino waveshield” and we have placed the order for this technology. While waiting he will research and accomplish all setup that is possible before the arrival of the technology.
Goals for Upcoming Month
Before the MGTR (February 15th), it is our goal to have a functioning device. This means our two technical leads on this project, Naomi Ray and Jalil Spearman will be leading the team towards integrating one of our “presence” sensors with the temperature sensor.
We will be working to integrate all of the parts that are currently functioning on their own. This is a very complex task, which is why we have scheduled another Saturday build day for February 4th.
The overall logic of our product is outlined here:
If ignition is on, device is off.
If there is an adult in the vehicle, device is off.
If the ignition is off and there is no adult in the vehicle, device senses repeatedly every minute for an hour.
If the device doesn’t sense anyone, device turns off.
If the device senses someone, SMS is sent.
If no one responds to SMS, SMS is sent again.
If no one responds and temperature rises to 80 degrees F, audible alarm is triggered.
If no one responds to audible alarm and temperature rises to 90 degrees F, GPS notification is sent to authorities.
Our goals that we have set for the month of February, include our sensor teams going forward with our testing and combining what we have completed so far. We are planning to have a three tiered alarm system for our invention, in order for the system to be complete we need to operate and interface the separate components for each level of the system.
On our team, communication is key! Within our testing teams, we have had all of our sensor groups that we have shaped; (GPS, Infrared, Motion, Thermal) come together and checked each other and verbally shared their progress, to make sure we are all on the same page. Also we have been diligently working within our teams to overcome obstacles with the sensors, like ordering unnecessary pieces, also making any adjustments to insure our sensors are functioning. We have checked in once again and shared out to the whole team what we had been doing and what our next steps are as a team. Since we can’t move on without a sensor we had make sure that all the sensors do work.
If a teammate needs assistance and another teammate has completed their task we will work together to make sure their daily goal is complete.
Since we have a lot of programing that has to be done we called in a professional that will help with us with a engineering techniques .Ms. White advised we contact one of her contacts who works as a software engineer. He gave us many new considerations to think about. Mr. Eric Ayers, has spend countless hours assisting of with our sensors and arduino programming. We all asked questions, in attempting to understand how to gather feedback on our product, so that when we connect all of our sensors, we will have the most efficient outcome. We have constructed diagrams of the data onto paper that will be easily accessible and will help guide us in our design.
While we are looking forward to getting feedback from the public on our design, the MGTR is right around the corner and we have a lot of work to do before the event!