Recently, a traffic accident that killed three people once again triggered a discussion on the safety of new energy vehicles. Rear-end accidents, bodies that caught fire, doors that could not be opened from the outside. Behind the triple attack of questions, it is necessary for us to figure out what kind of level of safety of new energy vehicles is. and how to avoid the above problems through technical means.
AEB is not a panacea, the oil circuit can also catch fire, but the door can't be opened?
First of all, once again, active braking (AEB) is not a panacea. The effect varies from car to car, to the objective environment (weather, etc.), to the speed of the vehicle, and even to the driver's operation. Each brand's function and tuning ability of AEB is different, which is easy to understand, and in the PPT of major manufacturers, this is also the part that focuses on publicity. However, the test effect is not equal to the actual application effect, rain, stains, fog, dark light, etc., will affect the judgment of the perception hardware, resulting in decision-making differences. At the same time, the volume, velocity (or stationary) of the detected object will also affect the system's judgment.
Even if the perception hardware such as cameras and lidars work under normal working conditions, and the system makes the correct judgment. However, the driver's actions also affect the results of the operation. For example, the driver actively intervenes in the brake, but releases the pedal, hits the direction sharply, slams on the accelerator (the wrong accelerator pedal), etc., all of which will cause the system to misjudge. To put it simply, AEB still guarantees that the driver's will is the first priority in vehicle operation. Behind this core value, the driver's personal quality, operation proficiency, etc., in case of emergency, are still the core factors affecting vehicle safety. To put it more directly, the driving assistance functions, including AEB, are all based on "assistance". You're just one more hidden multi-purpose co-pilot, but you're still the driver.
After emphasizing this old concept problem, the next problem of vehicle fire also needs to carry out a "literacy". It is true that the hidden danger of fire in the power battery part of new energy vehicles can always touch the sensitive nerves of users. However, for plug-in hybrid (extended range) models with circuits and oil circuits, the fire hazards of the oil circuit part cannot be ignored. Obviously, in the collision of the M7, the fire started in the front engine compartment of the vehicle. Regardless of the fire or the location of the fire, it has nothing to do with the power battery part at the chassis. However, this also reminds the majority of users that the safety focus of plug-in hybrid (extended range) models cannot only be placed on the circuit part, and ignore the most traditional oil circuit part.
Whether it is AEB or a fire after a passive collision, it is actually a problem that cannot be avoided in an absolute sense for the vehicle. In this way, the pressure is given to the last link of all problems. That is, after encountering an inevitable safety incident, the most precious life can be successfully saved. In other words, the fact that rescuers can open the door smoothly from the outside of the car is a safety detail that cannot be missed. But in this case, it is clear that the rescue operations revolve around the more complicated, time-consuming, and inefficient smashing of windows. So why can't the door be opened, and how can you technically guarantee that the door will be opened?
Hidden doorknob + range extender, don't carry the pot?
It seems to be a small door handle, but in fact, the operation logic is no different from high-end intelligent driving. It all needs to go through three steps: perception, judgment, and operation. Then, if the car door cannot be opened, it means that at least one of these three steps has gone wrong. First of all, perception is definitely fine. This is also one of the reasons why the airbags are repeatedly emphasized in preliminary reports, including officials, to open normally. The collision sensor is the source of everything, and only when it is successfully triggered can the airbag be deployed and the door will be automatically unlocked. For example, in some accidents, there is a problem of "missing the collision", that is, a very small range of overlapping offset collisions, resulting in no triggering of the sensor signal. The solution is simple, with more collision sensors, and more scientific algorithms.
So, since the collision is perceived and the signal is transmitted, why the door still can't be opened? There are two possibilities, the first is that the door is deformed by the collision, resulting in the failure to open smoothly. But this can basically be ruled out, because the accident hit the front of the car, and the door position is basically intact. In addition, the rescue operation has completed the broken window, but the door is still not opened, and it is obvious that the low-voltage part of the vehicle has most likely been powered off. Yes, even if it is a plug-in hybrid (extended range) and pure electric model with a large-capacity battery, its lights, locks and other parts are still low-voltage lines. In other words, the 12V battery, which is no different from the fuel car, is responsible for the power supply. Although many new energy vehicles, including the M7, have begun to transfer low-voltage batteries to the trunk part. However, this does not mean that the low-voltage electrical part of the vehicle will not lose power in the event of a high-intensity collision.
Even in this case, the electronic door lock can still be unlocked, because the door lock module is also equipped with capacitors. Theoretically, the capacitance of the door integrated circuit can still support dozens of opening and closing actions when the battery is empty, or ensure that the door can be opened smoothly in a short time. This redundancy is obviously sufficient for rescue after an instantaneous collision. It is true that the hidden door handle increases the workload of the door motor. But returning to the accident itself, it is better to think about whether the door did not receive the instruction to open it at all, rather than saying that the door could not be opened. To put it more closely, this means that the signal may have been broken off after it was transmitted to the airbag and then to the door.
The reason for this speculation is that in traditional automobile manufacturing, each operating unit corresponds to an electronic control unit, that is, an ECU. However, as the number of functions in the car increases, so does the number of ECUs in the vehicle. For example, the collision sensor gets the information and needs to pass it on to the airbag sensor and the body controller. The former will make the judgment that the airbag will be deployed, and the latter will make the judgment that the door is unlocked. In order to string together so many ECUs, the vehicle needs to be connected to a bus. However, the bus architecture results in long wiring harnesses and limited communication bandwidth. Especially under the trend of new energy with increasing functions, the upgrade and iteration potential of the bus is average, and the daily work pressure is too heavy. What's more, multiple ECUs are also passively damaged in the event of a collision, or the wiring falls off.
A patched solution is to design a hard-wired communication for some core ECUs, such as door controls, in addition to the bus. It is also possible to lay out the route in such a way as to avoid areas that are prone to collisions. In this way, even if the master control part is damaged, or even if some of the doors are deformed, there will at least be "surviving" unlocked doors. It's just that the cost of this solution is also visible to the naked eye. Hard-wired connections require additional wiring harnesses and less space-friendly layouts. These are not easy for new energy vehicles, which are becoming more and more functional.
Therefore, the root cure is still to think from the trend of new energy, that is, more in-depth integration. Obviously, the functional richness of new energy vehicles is subversive. So it's understandable that the number of control units is increasing rapidly. However, after all, there is an upper limit to the logic of traditional automobile manufacturing, and if you want to break this upper limit, you have to summarize the control domain. That is, to create a "middle layer" to be responsible for intelligent driving, cockpit, three electric and other functions, so as to accelerate the transmission of information and reduce the use of wiring harnesses. The end result, of course, is to go straight to a centrally integrated architecture. That is, by using the computing power and algorithm advantages of new energy vehicles, all kinds of perception hardware aggregate information to a "brain", and all decisions are also made by this "brain". The result is higher communication efficiency, as well as lower latency, power consumption, and collaborative stability. Even physically, this central control "brain" can be better protected from damage in passive collisions.
Write at the end: traffic safety has never been a topic of avoiding medical treatment, and even back to the era of horse-drawn carriages, there will be safety problems. What matters is how to reduce safety hazards and successfully save lives when accidents are unavoidable. New energy vehicles have brought users an unprecedented experience of driving and riding functions, and in terms of safety logic, new energy vehicles do not seem to have fully fulfilled their talents. The so-called "oil to electricity" was once only used to target the backwardness of the three-electric system. In fact, in terms of safety, it is also a more advanced way to solve problems by comprehensively using electrification thinking to build new energy vehicles.