New breakthroughs in machine learning will reduce petrol consumption
The introduction of digital technology and telematics systems has completely altered what, until not too long ago, we considered a mechanical vehicle. Today, our cars have become digital hotbeds with computerized systems controlling a vast range of maintenance, driving and safety systems.
This revolution, which began with satellite navigation systems, usage-based insurance policies and progressed to adopt preventive maintenance alerts, e-safety and sleep alert systems, and cars that self-adopt to different driving conditions, is now introducing automatic braking systems and a range of other spatial sensing systems that continue to improve our safety and that of others around us.
Moreover, ever since their introduction, telematics systems have focused on how to lower the costs of insurancepremiums (usage-based insurance), diminish maintenance costs through preventive alerts and reduce petrolconsumption thanks to the information received in real-time and processed by state-of-the-art satellite navigation systems.
Advances in Artificial Intelligence
As we advance towards a world of self-driving cars and the ITC and telematics systems in our vehicles grow increasingly more powerful, the benefits of machine learning will be applied to a range of new features and systems.
One interesting application of machine learning is the development of systems that will allow users to consume less petrol and reduce costs related to vehicle operation. While this has, to a certain extent, already been begun with the introduction of hybrid and electric cars, the truth is that these new vehicles need increasingly complex systems that can make the optimal choices concerning whether, at any given moment, the vehicle should be powered by its batteries or its fuel.
How Do Intelligent Transportation Systems Work?
The recent progress made in the application of artificial intelligence and the increasing availability of big data can now also be employed to study increasingly short journey segments – and not only yours, but that of any connected driver – and develop a database of specific driving conditions (road inclination, conditions and location, as well as weather, time of the day, traffic, etc.), as well as feed a machine-learning system that will allow cars to learn “from experience” whether to employ electrical energy or petrol (or other fuels) in given conditions.
Unlike a typical satellite navigation system that is typically used to reach new destinations, an experience-based system would be always-on, collecting and digesting information that will allow the system to rapidly determine the optimal power source and vastly improve our savings.
How Can I Save Money on Petrol?
As Researchers Qi and Barth at the University of California – Riverside report in “IEEE Transactions on Intelligent Transportation Systems,” initial simulations on AI systems applied to fuel consumption demonstrate that such a system is 10.7% more efficient than conventional ones.
Moreover, if the control system knows in advance that a stop will be necessary at a given point to recharge the batteries, the system can fully exploit the vehicle’s battery potential on each journey segment. In this case, the average fuel savingincreases to 31.5%.
Clearly, this would be a major bonus not only for private car owners, but also for fleet managers and all on-road transportation services and networks. We all like to save a penny and it now seems likely that before our cars will begin driving themselves, they will begin saving us money.