When we think of connected cars, we think of cars cruising the highways without a driver and then safety is inevitably the first thing that pops into our minds. In the midst of all the high-tech advancements, our well-being will always be the number one concern and for connected cars, it means safety features come first, before any other built-in functionalities.
Safety has always been an issue with vehicles, considering that the United States alone experiences over 10 million auto accidents each year.1 The first generation of safety features in automobiles was purely passive, involving seat belts and structural design. Then, as technology advanced, automotive industries enhanced their safety standards, installing airbags and distance detectors to warn drivers of potential obstacles. Now, with the futuristic technology of self-driving cars, we finally have proactive, built-in safety features such as the Advanced Driver Assistance Systems (ADAS).
However, ADAS isn’t a new technology. ADAS was created with the intention of keeping cars, passengers and roads safe, leveraging technology to minimize collisions and enhance traffic rule adherence and roadside awareness. It contains an umbrella of features for collision prevention and traffic regulation, incorporating software to regulate cooperative lane changes, merging assistance, collision warnings, speed detection and more
In fact, many cars on the road today already have ADAS-enabled features that are designed to alert the driver of potential problems. More advanced cars currently on the road even have embedded features that allow the car to take control in emergency situations. Yet for self-driving cars, ADAS is taken a step further by allowing vehicles to communicate with their surroundings on a whole new level – through vehicle-to-vehicle and vehicle-to-infrastructure communication.
Proactive safety technology for connected cars
Basically, ADAS features for the next-generation of connected cars are what will enable vehicles to “talk” to each other. Two of the sensor-based technologies are perhaps the most exciting features of ADAS – Vehicle-to-vehicle (V2V) & vehicle-to-infrastructure (V2I) and Vehicular Ad-hoc NETworks (VANETs) – these would be the top technologies in the intelligent transport system which make safety regulation possible for connected cars.2 Simply put, these are the underlying safety features that allow cars to communicate with those within a 300 meter radial distance in front, behind and beside.
What type of data do the connected cars transmit?
Via a Dedicated Short-Range Communication channel (DSRC), vehicles can detect each other’s’ presence, “lock-in,” and communicate critical vehicular information like size, positioning, acceleration, brake status, steering angle, turn signal status and more. This is done through complex data aggregation algorithms, where a vehicle collects its primary data records and compiles them into aggregated clusters or frames, ready for broadcast to nearby vehicles.3
How does vehicular data address the issue of traffic safety?
Vehicle-to-vehicle communication gathers other cars’ data and uses the information to avoid traffic related collisions via built-in ADAS safety functionalities, many of which may already be implemented in today’s cars (like the forward collision warning feature). But what is different with connected cars is that the real-time data it collects from nearby infrastructures and vehicles, will allow the vehicle to make human-like decisions through existing features, including, but not limited to, the following:
1. Intersection movement assist
This functionality detects multi-way congestions from data regarding the movement of other cars, to decide whether it is safe to proceed across an intersection. This is especially useful in situations where large structures block cars from seeing one another physically, but radio data can still be transmitted and received regarding each other’s’ vehicular information, so they can determine when it would be safe to move.
2. Left turn assistance
Just as the human driver must look at oncoming traffic when turning left, V2V will detect other vehicles in front of it, travelling in the opposite direction from directional and speed data aggregation, thus helping it decide if and when it is safe to make a left turn.
3. Emergency electronic brake light
The emergency electronic brake light functionality detects if other V2V-enabled vehicles are decelerating, thus prompting it to use the brake, this functionality will work even if the view is obstructed or weather conditions are foggy as well as if the stopping vehicles are not in its direct line of sight.
4. Forward collision warning
This feature detects and avoids impending rear-end collisions with other vehicles in the same line and direction of travel from within 300 meters, a distance safe enough even at high speeds. Additional sensors like single radars and single cameras are usually used as added complements to detect out-of-range stopped vehicles.
5. Blind spot warning and lane change warning
When the vehicle senses that its blind spot zone is occupied by another vehicle travelling in the same direction, this feature will advise against or halt lane changes. Though the accuracy of the algorithm sounds a bit mystifying, especially at incredibly high speeds, the feature seems to work extremely well in spite of concerns. Google and Delphi autonomous cars recently demonstrated its effectiveness in Silicon Valley- unintentionally.4 Delphi Audi Q5 was attempting to change lanes when Google’s prototype moved into its intended spot, and according to the Audi’s passenger, the smart car took the predicted measures to delay its lane change.
The Electronic & Safety Division of Delphi praised the “smart move” as “…a good example of how our car leverages complex software algorithms to assess its situation, much like a human driver would. Automated vehicles need to work through a decision via a highly complex network of technology and data processing to calculate the correct move. Our car did exactly what it was supposed to.”
Not just their car. All healthy cars equipped with the same V2V technology would do exactly what they are programmed to do in order to meet the expectations of safe driving.
The vital safety features of self-driving cars includes their proactive interaction with roadside infrastructures, adherence to traffic regulations is key to staying accident-free and this is where vehicle-to-infrastructure (V2I) technology of the ADAS comes into play.
Communicating with infrastructures
Giving smart cars the ability to sense and follow traffic regulations is a critical aspect of keeping roads safe. Just like passengers and conventional drivers, cars need to cooperate not only with other cars, but also roadside infrastructures such as traffic lights and road signs, so all cars can stay safe. Vehicle-to-infrastructure (V2I) functionalities allow cars to do just that.
Through the sensors embedded into self-driving cars, vehicles are able to communicate with sensors embedded into roadside infrastructures as well, through direct, short-distance communication, in a similar fashion to V2V.
Advanced features include:
- Red Light Violation/Stop Sign Violation Warning
- Curve Speed Warning
- Reduced Speed Zone Warning
- Spot Weather Information Warning
- Oversized Vehicle Warning
In many cases, V2V and V2I come into play in a self-driving car’s everyday driving decisions. As an example, when multiple cars approach a busy intersection, they will first sense if they have the right of way according to appropriate infrastructures like stop signs or traffic lights and then sense oncoming traffic to determine whether they can safely proceed. In a more complicated scenario, a car may have a green light with intention to turn left, it then needs to analyze the “data packets” it received through ADAS to make sure that the immediate light is green, via V2I, and approaching cars in opposite directions will give it enough distance and speed to turn, via V2V.
The Final Drive
With much of the road’s future safety directly dependent upon V2V and V2I safety features, ADAS software’s interoperability and scalability is essential to ensure OEMs developing autonomous cars see eye-to-eye or at least make their vehicles do so, in the face of evolving connected car developments. Automotive industries aiming to keep their cars and passengers safe can seek leading automotive platform experts for an end-to-end package to help reduce costs and shorten time-to-market, thus fueling that competitive edge.
Self-driving cars may still be in their infancy, but advanced driver assistance systems are becoming safer and more reliable. Of course, in this day and age, nothing autonomous is foolproof without millions of miles behind it. We still need to put connected cars’ safety features to test. But as connected car technology gains more credibility with more mileage and testing, and as companies at the leading edge of technology work with Altran to perfect them, we can roll out these self-driving cars with greater peace of mind.