We’ve heard about robots performing human jobs and chatbots that are programmed to reply certain customer messages, self-driving cars are just another step ahead to serving the human race better.
For self-driving cars to be fully integrated into everyday human routine, they need to have a system which can intelligently understand the surrounding environment, like complicated traffic scenarios, moving objects, human movements, roadblocks, potholes, drive paths, lane markings or any vehicle passing on the road.
However, the only way this can be achieved is when they understand the outside world just like humans do. Moving on the road will require seamless communication with other cars, passengers and surrounding traffic participants, this is how they determine their exact position on the road and decide on the next move to make.
Below, Naijauto will explain in detail ways self-driving cars communicate with different subjects.
1. Vehicle to Vehicle Communication (V2V)
Currently, we’re familiar with the 4G LTE connections, but we would be experiencing faster broadband soon. 5G is also a type of mobile broadband that allows the wireless transfer of data from one device to another, at a much faster rate than the 4G LTE.
At its peak, 5G promises to be close to 1,000% faster than 4G LTE, which will make issues like high latency and long response times a thing of the past.
Now, the self-driving cars will be equipped with 5G broadband, which will give way to seamless communication from one car to another. In addition, this kind of borderless communication will give autonomous cars access to exchange information about their current position, route, and hazards on the road.
For instance, if two cars are driving on a single lane highway and the car in front, through its onboard sensors, detects a pothole or any hazardous road condition, that information can be put across to the car behind. The car behind becomes aware of which spot to avoid.
If we’re looking at other advantages of V2V communication, we have to mention the possible elimination of traffic. With a whole network of interconnected vehicles, traffic congestion could be alleviated because self-driving cars will know what’s already ahead, avoid the pitfalls and that way, will maintain a steady rate of vehicle flow.
>>> Self-driving car crime: Engineer faces prison sentence for stealing Apple's self-driving car schemes
How self-driving cars work
2. Vehicle to Infrastructure Communication (V2I)
Communicating with other vehicles is one thing and communicating with road infrastructures is another. Once self-driving cars are connected to a 5G network, they can also communicate with different infrastructure elements that make up the human transportation system.
Let’s use the parking system as an example:
While driving and deciding on the next move to make, it’s imperative that self-driving cars also plan to park in the best available spot. Therefore, information about available parking spaces can be transmitted over the air to a self-driving car through sensors that monitor whether a parking spot is occupied or not.
Once this information is received by the car, that space can then be reserved for that specific car. To enable easy access for other driverless cars, that reservation can be broadcasted over the cloud so others can be aware it’s occupied.
>>> Highly recommended for you: 5 different levels of autonomy in autonomous cars
3. Vehicle to Pedestrian Communication (V2P)
The most important communication is with pedestrians, the vehicles need to be aware of pedestrians and their exact location at every second.
Most times, we leave home with one smart device or the other, our smartphones or any kind of internet ready device. This means that in one form or another, we are almost always connected to the internet. Many of these devices have the ability to use GPS to determine the exact location of its current handler.
With self-driving cars and smart devices using the 5G network, the current location of a pedestrian can be instantly relayed to a nearby driverless vehicle, making it aware of the pedestrian’s whereabouts at all times.
Autonomous cars collect data with the help of various sensors fitted in them like cameras & LIDAR
>>> Keep yourself updated with Naijauto: Ford develops 2 new technologies that allow cars to communicate with road users
This stage of connectivity with pedestrians will allow driverless cars the opportunity to react dynamically to the position of a pedestrian with collision prevention measures like braking and automatic steering.
In addition to the V2P communication, there will be a centimeter-level accurate, digital 3D representation of the physical world on a map. The data on the map will serve as the main source of guidance for autonomous vehicles.
Serving as an integral part of the system, the HD Maps (High Definition) also bring functions such as high-precision localization, environment perception, planning and decision making, and real-time navigation cloud services to driverless cars.
4. How do maps help self-driving cars communicate?
Quite different from the regular map, a map that supports autonomous driving constantly detects, verifies, and updates changes that happen around the world. It is created in four simple steps. They include Collection, Aggregation, Creation, and Publishing – let’s take a quick look at how these four processes help self-driving cars communicate effectively.
Autonomous cars collect data with the help of various sensors fitted in them like cameras, LIDAR (Light Detection and Ranging) and radars. Afterward, the data is transmitted back to the cloud.
This crowdsourced data can be anything, as long as it’s crucial to decisions a self-driving car makes and how it functions in its entirety. The information could be lane closures, barriers, road signs, and pavement markings. You should be aware that this sensor data alone is not entirely correct, neither is it ample enough to completely take an actual driver out of the picture.
Driverless cars come in different shapes, sizes, travel different locations and while they all have sensors, they are placed at different places. With all these variations, they perceive objects differently and capture data differently as well. Thanks to the machine learning algorithm, data captured by different cars are fused together so that they can have accurate and unified features.
>>> By the way, Google launches travel mode map in Nigerian voice for cars & motorcycles! Check the news!
As soon as the data has been accumulated and accurate data has been generated, the map will be created. This is done with the help of refined algorithms and unique features. On the map, all accurate information about the physical world is represented, it helps determine the exact position of an object. This process is executed with advanced algorithms that take into account various data collected from cars, and then the feature for a map is created. For certain features, ten observations may be needed, or even a hundred, it all depends on when the algorithm starts converting the many features into one accurate feature.
As soon as all the data has been generated and the map has been created, it is then updated and published. To ensure accurate data transmission is taking place, only the updates that occur within the specific tile for the specific layer – the Road Model, HD Localization Model, and HD Lane Model, is sent to the OEM’s (Original Equipment Manufacturer) cloud and then, the vehicle.
With the tiled format, over-the-air updates can be sent in a more contracted package for efficiency and optimization of data exchanges. When a new feature is published, there may be a particular area of the road where enhanced sensor data is needed. For instance, a driverless car might not identify a stop sign due to certain obstruction. In such cases, the map will request that the next vehicle in that area takes a video of the environment, and then the system can better validate the data.
This happens through the SDRI (Sensor Data Request Interface).
>>> See why self-driving cars won't work in Nigeria: This is why self-driving cars are unfeasible on Nigerian roads
5G broadband will be equipped on self-driving cars.
Three-layer information system
When the map is published, it has three layers. Each layer provides detailed information that supports self-driving vehicles to connect and communicate with the external environment as well as other vehicles.
This is the road model that offers global coverage and helps vehicles to understand local insights beyond the range of its onboard sensors. Such insights include high-occupancy vehicle lanes or country-specific road classification.
This is the HD lane model for more precise lane-level detail like lane direction, lane type, lane boundary, and lane marking types. These details help self-driving vehicles make safer and more comfortable driving decisions.
This is the HD localization Model. This layer helps the vehicle localize itself in the surrounding environment and helps the vehicle to identify objects like guard rails, walls, signposts, and pillars. Furthermore, it uses the object’s location to measure backward and calculate the exact position where the vehicle is located.
Well, with all the technology staring right at us and waiting to envelop Nigeria, the fact still remains - driverless cars won't be at our service anytime soon. The good news is, you can always purchase other befitting cars in Nigeria.