Introduction

With autonomous vehicles becoming more prevalent on today’s roads the ability to optimize traffic flow by adjusting vehicle software is more in demand than ever. Our team is developing such an algorithm that will be distributed to all smart vehicles on a network. Once activated, the vehicle will automatically take in a number of different inputs such as current traffic conditions and provide the car’s navigation system with the optimal route to reach its destination.

In order to reach maximum vehicle throughput of the network we’ve decided on the following areas of focus for the algorithm:

• Primary congestion avoidance
  • Using machine learning (see Smart Car Approach below) our algorithm will be able to understand which routes will lead to the greatest amount of congestion and react by instructing the car how to avoid these situations.
• Secondary congestion avoidance
  • Here we increases efficiency on multi lane roads without changing a vehicle's current route
  • Our algorithm will do an analysis of all vehicles ahead of the current car (within a given range) and determine the optimal lane to reside in