Range-Finding Rear-Collision Accident Prediction and Warning For Motorcycles with Integrated Signal Vest

The project at hand

There is a lot of technology that helps car drivers avoid accidents. There are lane monitors that will detect an undesired lane change then steer the car back into the lane, blind-spot detectors that display a warning when someone is driving where a driver can't see them, and environment mapping sensors that are connected to accident prediction systems that deploy automatic braking to prevent accidents. Motorcycles do not have any of these safety features. Granted there are limitations for what a motorcycle accident prediction system can do. These systems are able to warn the rider of whatever it is that they detect, but cannot perform an automatic correction. Since motorcycles are two wheeled systems, any unexpected change in braking or steering can cause an accident instead of prevent one. If the front brakes were to be automatically applied during turn, the physics of a bike would force the bike to enter a low-side slide. Unexpected steering adjustment would change the balance of the bike and would disrupt the riders’ own balance, leading to an accident. However, an accident prediction system would not be useless for a motorcyclist. Warning a rider of an impending accident can give them much needed time to follow an escape path and get out of trouble.

It was decided that the type of accidents to be predicted are rear end collision accidents. By rear end collisions it is meant of the situation where a vehicle behind the motorcycle collides into the rear end of the motorcycle. The choice to work on this issue comes out of feasibility and pertinence. The feasibility of the project comes from the comparison with the correlative accident prediction system. For a front-end accident prediction system, (motorcycle running into something,) the situation of cars suddenly changing into the motorists’ lane would have to be a part of the system. This limits the sensor type strictly to a camera system coupled with image processing. The comparison to the rear-end requirements of a distance measuring sensor shows that a rear-facing accident prediction system is much more feasible to be completed in the working situation. A rear-facing accident prediction system is also more important to the community. Motorcyclists have significantly larger issues seeing the road behind them than in front of them. Motorcycle helmets provide strong protection in an accident, but take away a fairly large chunk of peripheral vision. Combining this fact with the largely unknown fact that motorcycle mirrors vibrate with the engine creates the conclusion that motorcyclists have a hard time seeing behind themselves. An accident prediction system that acts as an additional pair of "eyes" in the section that motorcyclists have trouble seeing would be a monumental asset to motorcycle safety.

The project at hand is a proof of concept design of a simple accident prediction system can be made cheaply and implemented onto a motorcycle. The decisions to make the project proof of concept came from the need to simplify the project to a manageable level that would be able to be completed without industrial simulations. A full design would involve applying the system to the motorcycles' built in Engine Control Modules (ECM). Due to the fact that consumers are restricted from reprogram ECM's, this system will be limited to proof of concept. The accident prediction system will utilize a Time Of Flight (ToF) sensor that will determine the distance between the motorcycle and the vehicle following. Using measurements taken at timed intervals, velocity and acceleration of the following vehicle can be calculated [1]. With these, a set of calculations can be made to determine if the following vehicle is projected to hit the motorcycle ahead. The second part of the project is to develop a system that will warn both the motorcyclist and the driver of vehicle behind and the rider to the detected accident. The rider will have a console based warning system that will indicate when a potential accident has been detected, while the following vehicle will have a warning that comes from the rider. To warn the following vehicle, a vest that has an integrated LED vest that will, in emergencies, display a warning signal. Pending consultation from the police to determine if the color will be allowed on the road, the warning will be a flashing exclamation mark in orange. In addition to the warning, the LED vest display the break and turn signals for added visibility and safety.