A few weeks ago, we did an article on data in the world of transportation. In this article, we will go into more detail about data related to the automobile

The autonomous car, a story of data

In a few years, autonomous cars will be driving on the streets of our cities. Data collection is at the heart of this subject of the autonomous car. Indeed, collecting data will allow to anticipate as much as possible all the situations that may occur. 

Of course, we need to feed this data with machine learning models in order to make predictions. And the errors of these models must be close to zero to have an acceptance by the population. 

Although this technology is progressing, there are still some failures. For example, a pedestrian is walking on a sidewalk with a stop sign on his shoulder. The vehicle then stops, starts and stops again, seeing a stop sign all the time. Can these cases, although very marginal, be avoided in the future? Opinions differ. 

Predictive maintenance: prevention is better than cure

Putting the right sensors and collecting the right data will allow to anticipate the fragility of the parts and to change or reinforce them before they break. A part that breaks can jeopardize the solidity of the car, will surely create a breakdown. If we reinforce or replace the part just before its breaking point, we will avoid any inconvenience. Predictive maintenance will initially concern transporters but should gradually be deployed for all types of vehicles. 

In addition, the data collected by these sensors is analyzed by the engineers of the various manufacturers to make continuous improvement and thus improve the quality of their vehicles. If there is enough data collected, the driving data can be used to train and improve the autonomous driving algorithms. 

Also, note that much of this data is brought to the driver's eye. No need to measure tire pressure or read the oil level, the information is directly present on the dashboard. Find the nearest gas station at the best price? Your car can help you. 

Driving assistance 

The data allows various levels of driving assistance. The first level of assistance is the reversing radar. However, more advanced versions exist. When you have a danger on a side of the car, your steering wheel can start to vibrate on this side to signal you this danger and allow you to anticipate it well. 

We can also imagine applications where on a weekly or monthly basis you receive a report giving you the points you can improve for your driving. For example, to help you anticipate which trip will make you consume the least depending on the time of day you drive.

Connected boxes: rewarding good drivers

The method does not seduce many French people for the moment: pay your insurance according to your kilometers and ... your driving behavior. Indeed, for many people, this device is too intrusive. For the moment, it is more appealing to young people who do not drive much, potentially allowing them to reduce the cost of their insurance. The device is usually inserted in the glove box and does not require any installation and is battery operated. 

Autonomous vehicles, predictive maintenance, continuous improvement of vehicles, user assistance, connected boxes are all use cases of data in the automotive industry. These use cases are still in their infancy and will continue to develop, more and more strongly, in the years to come! 

‍