Cities, public administrations, and transports are encouraged to make better and safer mobility systems where traffic is based on IoV technology that uses information as environmental pollution, particulate matter, weather, noise levels, and the influx of people to issue alerts and recommendations automatically, as well as make emergency decisions in traffic and vehicle management.
In order to monitor these values, visualize them, and decisions making, HOPU has substantial experience working on providing data-driven simulations and a predictive analytics workbench for experimentation with what-if scenarios and the design of proactive measures, e.g. for managing Dynamic LEZs (D-LEZs) and enforcing automated workflows. D-LEZ Design and Interactive Maps: Currently, most of the LEZs are static with specific boundaries and operate in a time- and seasonal-varying way. The D-LEZs functionality, empowered by an intelligent DSS, will enable modeling and eventually enforcing specific measures based on predefined conditions, such as automated Dynamic Charging Schemes. The boundaries of the zones will be dynamic to enable the adjustment of the intervention areas based on accurate measurements of the environmental conditions. The notion of the D-LEZ sets the foundation for the application of new types of zones (e.g., Zero-emission, vehicle type/model-based, etc.) and the expedient alignment of urban planning to specific regulations. Parameters that could affect the D-LEZ perimeter are indicative:
a)Real-time environmental conditions collected by the environmental sensor nodes in the periphery of the targeted area
b)Meteorological conditions and forecasts
c)Environmental situation of neighboring areas
d)Extreme events with environmental impact (i.e. fire accident in a chemical factory in the area) or particular circumstances (i.e. roadworks, holidays, events, etc.).
In image 1, for example, in a hypothetical scenario, if the targeted area (orange) is over-polluted for a couple of hours, the D-LEZ will block the entrance of LGVs in specific entry points (red dots).
Image 1. AQi monitoring by Grafana visualization. D-LEZ perimeter.
HOPU utilizes a combination of traffic simulation models, focusing on microsimulation, operational level, and agent-based models. HOPU significantly improves the analytical capabilities of the sophisticated econometric models with a multitude of quantitative and qualitative tools, mostly with the use of detection nodes (for calibration in a very detailed and organized manner, improving the forecasting ability of the model and allowing for a detailed simulation of the network) to explore and model human behaviour related to transport. By precise modelling and accurately depicting travel behaviour,
The level of detail provided by state-of-the-art models is very useful when considering small-scale interventions such as dynamic pricing in specific zones, AQ and noise modelling, emissions charging and other policy interventions. Most traffic simulation models are equipped with GHG emissions models (such as HBEFA, CHIMERE, and POMI). COPERNICUS services will be used in AQ modelling, providing input to the DSS regarding emission-based charging schemes.
The datasets provided for the C-ITS Roadside Station Use Case will contain regular operation messages and malicious messages which will be received by the Road Side C-ITS station (R-ITS-S) over the ITS G5 interface (based on 802.11p). This information will help the SerIoT anomaly detection module and the honeypot to detect malicious messages. C-ITS messages like Cooperative Awareness Message (CAM) and Decentralized Environmental Notification Messages (DENM) are the two mainly used message types transmitted from a vehicle to the infrastructure within C-ITS, the dataset will focus on these two. Normal correct messages and malicious messages will be part of the datasets.
The Datasets will be generated with Vector CANoe Software v11 and v12 (after its release in Q2 2019), using the built-in Car2X features. The messages will be from the Vienna Living Lab and use data elements and corresponding content specified within C-Roads, which is a joint initiative of the European Member States and road operators for testing and implementing C-ITS services. (https://www.c-roads.eu)
The datasets will contain CAM, which are sent out every 10th of a second and DENM, which are sent out when certain events (e.g. stationary vehicle) occur. Both of these message types will be signed with valid certificates as standardized in ETSI TS 103 097 v1.3.1, allowing them to be within the C-ITS trust domain. Under normal conditions, CAM and DENM will have the correct information and use all the mandatory and optional data elements correctly as described in the C-Roads Document on C-ITS Infrastructure Functions and Specifications.
Datasets containing malicious information will contain corrupted CAM and DENM. As the messages contain more (or in some cases less) information than specified, their length will vary accordingly. If a message differs in size from what is expected under normal conditions, it shall be treated as an anomaly.
Malicious Datasets can also be filled with messages consisting of incorrect information or code, which could be used to exploit R-ITS-S's vulnerabilities. Although detecting such anomalies may be outside the scope of network anomaly detection, insertion of incorrect information is a major concern for C-ITS and could be detected by checking for inconsistencies when comparing messages.
Image 2. Smart Spot IoT system
Currently, HOPU is immersed in projects related to the monitoring of air quality and sustainable mobility with entities such as the city of Tampere with the company Capelon, the University of Granada (Spain), and in the process of Pre-commercial Public Procurement of the Community of Madrid. You can find all the information about our AQi technology and real-time monitoring at www.hopu.eu