As we already wrote, IT is all about data. Mining data is like to extract gold with the difference that data are useful, gold is simply rare.
The whole world is a data mine, not only Internet, books or documents. Please, pay attention about the fact that sources of data and the media used to share information are often mixed up and this is really a great logical pitfall. The data (well collected and artfully analyzed) describe phenomena, any kind of media only shares the information. When the media start creating facts, we became all George Orwell characters.
In order to collect data to make customers life easier and improve safety, one good idea is to use transportation as moving probes. Obviously, the first step is to equip cars, trains, buses and, why not, bikes, with some kind of data collector unit and a small network appliance. Modern cars, in fact, already are equipped with several sensors to feed their internal computer units in order to manage safety equipment and other automotive world aspects. When a car is on a road, it measures external temperature, it knows if it is raining or if the road is icy. The car knows its speed and, if for some reason, ABS, Anti Skid, ESP units are activated, the fact can became really interesting for a control traffic center because if several cars have the same problem in the same road, may be that something interesting is happening.
European HeEro eCall project, starting from 2015, will oblige car makers to install on new brand vehicles an on board unit able to activate
an electronic safety system automatically calling emergency services in case of a serious accident. Even if you are unconscious, the system will inform rescue workers of the crash site’s exact whereabouts, and the rescues will be on its way within minutes. The system, which has been baptised “eCall”, is going to work all over the European Union. (read more here)
In Telecom Italia Tilab, I’m the leader of a research group that is studying a set of applications/procedures to improve the eCall basic safety features collecting the data using vehicles and their on board sensors as probes and activating some logical inferences in order to offer a better rescue service and a general management of transportation network. The project is called e2Call to evidence that is an eCall system improvement. The project will be, hopefully, presented in the next 10th ITS European Congress in June 2014, in Helsinki, Finland.
A small presentation follows:
an eCall enhancement proposal
based on 4G telecommunication network
and inference algorithms.
In 2015, countries defined as “pilots” in the HeEro European project will oblige car makers to install, on new brand vehicles, an electronic safety system automatically calling emergency services in case of a serious accident. We want to manage the car sensors and the processor of the on board unit (OBU) to assemble an enhanced eCall that feeds an inference engine to give an interpretation of the accident dynamics to provide more accurate rescues. Using 4G network, cars will also share information about their positions and routes for advanced driver assistance systems based on on board data processing and centralized inference engine advises to avoid collisions. Cars themselves could be used as moving sensors to determine dangerous road conditions and traffic congestion. The research is partially founded by The Autonomous Province of Trento government and developed in collaboration with CRF (Fiat automotive research centre) and Department of Information Engineering and Computer Science of University of Trento.
E2Call main topics
The inference engine
Automatically alerting rescues services in case of car accident is a great improvement in car safety world. On another hand, if the passengers are unconscious (and, often, it means that the accident is very serious), the PSAP operator receiving the call is only informed about car position and that something happened causing air bag intervention. Collecting car’s sensor data and sending them with a short time history of the evolution (i.e. last 10-20 seconds), an inference engine, based on a knowledge base, improved with experience, can formulate an hypothesis on accident mechanics and consequences. I.e. data coming from vehicle speed sensor, throttle position sensor and ABS intervention sensor can be combined to estimate the severity of accident and, in case of internal thermal sensors availability, the quick growth of temperature can suggest the presence of fire. The elementary information collected are sent via e2Call in addition to the minimum set defined in eCall MSD and elaborated by inference engine to assist PSAP operator in evaluating accident consequences.
Car are identified via vehicle identification number (VIN), so if dangerous fuel is used (like LPG) an active alarm can be sent to vehicles travelling in the accident surroundings. Identity of passengers on car during the accident can be sent also reading (via smartphone) the innovative Trentino Sanitary Card when boarding the car. So, the PSAP operator is informed about specific people sanitary needs recorded in Trentino on line Medical History TreC and appropriate rescues could be dispatched.
The initial accidents knowledge base is based on public available data reporting accidents, boundary conditions, car sensors values (if available) and accident consequences. At present time, global data aren’t available, so the experimentation will start using United States Census, SafeNY, UK Government Road accidents and safety statistics, NHTSA and other sources that we are still evaluating. In the future, e2Call improved knowledge base could be used for other car safety applications.
The 4G network role
The availability of a high performance and large band wireless network covering plays a strategic role in inter-vehicle communications. The on board units UICC of vehicles can share a multiplexed dedicated channel in the 4G cell allowing all vehicles to be connected without the installation of dedicated> IEEE 802.11p appliances and usefully filling available network resources. A precise study and experimentation on 4G network standard features, allowable bandwidth, RTT and latency will be reported to confirm the technical feasibility of a shared wireless network as media for this kind of service. The 4G cell extension assures the short range required to collect and broadcast information related to a limited region of space avoiding that information not related to vehicle position ad route be uselessly broadcasted to the vehicle.
The road safety information system
Using the same hardware required by HeEro project, in e2Call vehicles not involved in accidents can be used to anonymously send information. Data will be stored in a central database and unexpected events could be automatically analysed. I.e., matching season and weather/traffic reports, the fact that several car sensors signal the anti-skid intervention in a precise position could mean that something has changed road adherence requiring Road Maintenance Service intervention and an alarm broadcasted to vehicles in the surrounding area. In the same way, if an intersection causes statistically more ABS activation than others, it could mean that intersection is dangerous for lack of visibility and requires traffic lights or some different road signs. This kind of service makes an accident prevention policy possible.
In the same way, vehicles medium speed on a road could be related to time series data to be informed about a particular road congestion and to send an alert to vehicles incoming in the area.
A dedicated, non-distracting and easy to understand smart phone interface is going to be designed in order to safely broadcast high level information elaborated and strictly regarding the area in which the vehicle is traveling.
The advanced driver assistance systems
The availability of an on board unit with an high performing cpu, network connected to vehicles in the surrounding area could be a powerful tool for the activation of emergence, automatic intervention of brakes and/or emergency alarms.
In this specific case, network RTT and latency play a fundamental role. To be effective, an advance driver assistance system intervention should occur in less than 100 ms. For this reason the performance of the network platform should assure this kind of result, the position of vehicles has to be very precise and the on board unit has to be capable to elaborate a lot of data in the smallest time interval possible, probably adding some hardware to the minimum set required for e2Call standards to improve computational performance and geo location.
All this factors are going to be tested and measured in e2Call to proof the feasibility of the e2Call architecture in advanced driver assistance systems. It has to be considered that this section of the project is more related to car maker offer, to sensors and ADAS tool availability.