Recently couple of years we seen the problem of parking in metro cities. According to telegraph, an average driver wasted 2,549 hours for finding parking area. That mean it around 106 days. This is the static of UK, if we talking about the other countries like United state, India , so this statics is more. So google has bee launched smart parking in smart cities.
To solve this issue, google launched smart parking by the help of google map. Google start this features across 25 cities in United states. This features is not start for all cities or country. But if you are in live in those 25 cities. so you will now see the red parking sign on google map that indicate the limited parking availability to plan your trip.
The function of this features is quite different. It work with the data of crowdsource and relatively machine learning algorithm. I said it’s quite different because previously these type of features is depend on internet connected parking meter. Which provides the wrong data many times, due to invalid or illegal parking area. But due to the uses of Artificial intelligence, this time we probably gain the accuracy. And peoples easily book area on smart parking.
Machine Learning Application
Only google know the algorithm is how much complex. Because firstly google find out blank parking area. Then if any vehicle is come on the that area. So now google measure the false positive with measuring the time and location data. And ensure that the person is come with own vehicle or they arrived via taxi. Its prevent Fooling the system into thinking that parking space is available or not.
Now the next thing is that if any person is booked the parking area. So first google is calculate the distance and time. And find out how much time will take by person or that particular car to reached at parking. If the car is come via different or long path up to the parking. if the large difference is notice by application so it indicate the message is “more difficult to find parking area”. Due to this the application accuracy will be increased. Google also added other features such as variability based on time of day, date, and historical parking data to tune the model.
Advantages of this parking for google
- They not need to install extra information gathering sensor.
- It able to work with stored data and live data.
- Gain more accuracy.
- Peoples easily find out the parking.