And finally, here we have it - the Bike Kit. It's been a struggle getting it to this stage, but along with the other parts to this project, I'm happy with the result. Unfortunately, I haven't had the chance to 3D print and prototype the Bike Kit, or get any photos of it on my motorbike, but in my opinion (and why would I be biased) the renders do a great job of showing the intended product.
So let me walk you through it!
What Is The Bike Kit?
The Bike Kit is the brains behind the Blind Spot Detection Kit. By using AI technology, it recognises when the rider might be sitting in a blind spot, and sends this data via Bluetooth to the HUD to alert the rider. The Bike Kit sits on the front wheel fender (the front mudguard) to allow an unobstructed view of the road ahead.
The Bike Kit is designed to be aerodynamic and sleek, yet as unobtrusive as possible. I didn't want the kit to overpower the design of the bike, but to sit within it, doing its job while not being obvious that it's there. The whole Blind Spot Detection Kit is designed to be as unobtrusive as possible. Below is a picture of the entire product fitted to a bike.
The Bike Kit is relatively small, measuring 80mm wide X 120mm long. This size ensures it can fit onto almost every bike, and even most mopeds and scooters too. The Bike Kit attaches to the bike by using 3M tape, the same adhesive used for attaching Go Pros and other accessories to bikes and sports cars. This will give the user confidence that the kit isn't going to fall off and break mid outing.
The main body is manufactured of out of ABS plastic, making the kit robust enough to withstand any knocks and scratches.
How Does the Bike Kit Work?
The Bike Kit relies on a small, wide-angle camera that continually scans the road for any blind spot threats. AI technology uses this camera to recognise what a threat might look like, and only when one has been identified will it notify the rider. This ensures that no false positives are being highlighted to the user, keeping them in the safest possible road position at all times.
The Bike Kit will get its power from the 12volt battery on board the bike. To make sure the kit doesn't drain energy from the battery while the bike is off, it will also be wired into the ignition switch, only pulling enough power to send the summary email before turning off completely.
When the rider next turns the bike on, the kit will start pulling charge from the battery again, turning it on automatically. The kit will never take enough charge to damage the battery, as the technology within it runs at a very low voltage.
Unfortunately, due to working from home I never got the chance to build a fully functioning AI model. Still, to prove that technology is possible I made a person detection model which can be demonstrated from inside my flat.
Below I have attached a video to show how it works!
Inside the kit is a wide-angle camera, which is slightly set back into the body to prevent anything catching or damaging it which out on a ride. For prototyping purposes I've used an Arduino Nano as the camera and AI processor, but when in manufacture faster, more reliable hardware will be used.
Due to coronavirus, I haven't been able to build a fully functioning prototype of the Bike Kit, but I have developed a prototype using person detection technology instead. To create this, I used an ArduCAM 2mp camera and an Arduino Nano 33 BLE Sense. I've attached a copy of the schematics for this prototype below.
And there we have it! That's the Bike Kit. Overall I'm pretty pleased with how this part has turned out, despite all the challenges presented along the way. I'm pretty disappointed that I never got the chance to make it real, but given everything going on at the moment I'm still happy with what I've got. Thank you for reading!
Comments