It is hard to imagine that in the last 100+ years (2022), we have been able to populate Earth with 1.45 billion vehicles. With 8 billion people on the planet, 17.7% have a vehicle. With this many vehicles driving, we encounter many challenges to accommodate them. These challenges include scaling infrastructure for parking them, building more roads to handle traffic congestion, and developing sustainable transportation to reduce emissions. Each challenge gives birth to innovation for solving them. A key innovation on our roads is self-driving vehicles, which have the most significant impact on solving the growing population of vehicles on our planet.
A self-driving or autonomous vehicle can move safely without much human input by using sensors to detect its surroundings. There are six levels of driving automation, ranging from fully manual to fully autonomous. Most vehicles are at level two automation. The driver-interaction ratio needs to be one to two million miles per interaction to move to higher levels. The average consumer drives around 10-12 hours per week, but autonomous vehicles could operate up to 50-60 hours weekly, making them much more useful. They could even provide ride-sharing services, reducing the number of vehicles on the road.
A planet full of self-driving vehicles is our hope, but the road to getting there comes with challenges. They have data-hungry sensors to monitor their surroundings, generating up to 1 gigabyte of data per second. About 30 seconds of driving would fill up storage on a typical iPhone at that rate. If we want to achieve full self-driving on every vehicle, we will need to alleviate the following hurdles:
Collecting large sets of vehicle data at a daily rate is costly.
Collecting quality vehicle data is essential to train neural network models.
Processing relevant data to neural nets to continue to improve them.
The transition to fully automated vehicles on our roads is ongoing, but we face hurdles to achieving it effectively and safely. Collecting and processing vast amounts of vehicle data is crucial to achieving this goal. Self-driving vehicles could positively impact road safety, traffic congestion, parking, and pollution reduction. With more access to vehicle data sets, we can accelerate the future of automated vehicles on our planet in years rather than decades.