Project
| # | Title | Team Members | TA | Documents | Sponsor |
|---|---|---|---|---|---|
| 71 | Rear Collision Bicycle Warning System |
Gus Kroll Justin Davis Seongwoo Kang |
Xihang Wu | appendix1.ino appendix2.py design_document1.pdf final_paper1.pdf photo1.PNG presentation1.pdf proposal1.pdf |
|
| ##Problem: In the US, many cities and densely populated areas such as college campuses, do not have adequate infrastructure for bicyclists. This leads to an often unorganized and chaotic environment for cyclists to ride. While it is easy for most cyclists to manage this environment when they can see it coming, it is much harder to also be paying equal attention to cars, other cyclists, and objects approaching from the rear. Similar to cars, even with rearview mirrors and side view mirrors, most manufacturers are still including lane change sensors. The added value of this technology is that even though you can look at mirrors to see what is around you, it requires all of our attention just to manage what is going on in front of us and these sensors take on a lot of the burden of notifying us of potentially dangerous situations developing outside of our direct line of site. ##Solution: The goal of this product is to notify cyclists of objects approaching from behind so they can maintain focus on what is in front of them. The field of view would be approximately 90-120 degrees. This would cover objects directly behind and objects behind and to the side of the rider. This system should be low cost, able to be used on all types of bicycles. It is not intended to be a collision avoidance system, simply a warning system. The system will work using multiple ultrasonic sensors. Using a bit of hardware filtering to reduce noise and, some math and software to control object detection we should be able to generally ensure that we can identify when an object is moving towards the bike from behind at speed and give the rider a warning that something is approaching from behind via some combination of vibration in the handlebars/seat, LED’s and sound. ##Systems: Power - Battery, On/Off Switch Sensors - Ultrasonic Sensors (at least 3) Warning System - Noise Filtering Components, LEDs, DC motor (can deliver vibrations to handlebars), Speaker Control - Microcontroller ##Criterion for success To begin, in a controlled environment we want to be able to detect an object's location and determine it’s implied path (will assume linear movement for processing speed) at a distance >20 ft, and notify the rider. The next step would be to replicate the above results while maintaining a low false positive rate but increase the distance to 40-50 ft reliably. Then, again replicate this test in a moderately busy environment such as riding around campus, except on Green St. Green St will be the trickiest environment as there is the most noise to filter. Again the criterion for success in these more complex environments would be the same as in the lab environment. |
|||||