This week we introduce next AIthena consortium partner: MAPtm
MAPtm, founded in 2010 is a SME and service provider in the domain of Traffic Management. Apart from traditional Traffic Management, MAPtm explores new, innovative services based on for example data analytics, social media, Mobility as a Service and Cooperative Intelligent Transport Systems (C-ITS). MAPtm operates as an independent partner for road authorities, national and local governments, municipalities, and contractors. MAPtm is active in the complete chain of Traffic Management, from the design phase, including development of new smart algorithms, up to the exploitation and operation.
MAPtm has implemented its own Traffic Control centre from where Traffic Management Systems from different cities can be operated. MAPtm is active in national and international architecture and standards development for Cooperative, Connected and Automated Mobility (CCAM) deployment, partakes in several EU-funded research and innovation projects and frequently supports Rijkswaterstaat with their work in the Amsterdam Group, Car2Car consortium and C-Roads platform.
What is the role of MAPtm and the team in the project?
MAPtm is responsible for task 5.5 UC-4 AI-based Traffic management. That task focusses on the execution and evaluation of use case 4. To enable that use case MAPtm covers the needs and requirements of that use case in other parts of the project (Work Packages: WP1, WP3 and WP4).
In short, use case 4 focusses on:
(1) analysis of the response of AI-models to traffic management data, at strategic, tactical, and operational levels, and
(2) evaluate the impact of the vehicle behaviour because of the AI on the traffic system.
What are you currently working on in the project?
Artificial Intelligence (AI) is a new subject to public authorities. MAPtm intends to transfer the knowledge developed in AIthena to our clients and help them to improve their operational processes and traffic management systems.
MAPtm will focus on the demonstrator “AI Models and Traffic Management” (use case 4), related to utilisation and coordination of AI-based components at vehicle-level in transport-level applications (traffic management). The capabilities of the AI models will be tested and evaluated in scenarios that are particularly relevant for traffic management and road authorities. The decisions of AI models in response to the data are of special interest. The behaviour of the vehicle will be benchmarked against what is considered good behaviour as defined by the rules of the road. For evaluation purposes, causality together with transparency and its predictability is important for road operators and traffic managers to assure effectiveness of measures taken and information disseminates to manage traffic on their road network. Moreover, it will provide road authorities with a new framework to objectively assess AI capabilities in different traffic conditions, situations, and scenarios.
From your perspective, how do you see the contribution of the AIthena project to building trustworthy, explainable, and accountable AI-based CCAM?
The AIthena project will help to understand what variables have an impact on the behaviour of automated vehicles (AI-based CCAM) and, vice versa, what the impact is of that behaviour on other actors. The understanding of those relations, and the causalities within them, will lead to transparency between different actors. This can be between:
- Different systems (e.g., traffic light controller and vehicle, or between vehicles)
- AI-based CCAM developers (OEMs) and road operators, vehicle approval institutes, and other relevant organisations (e.g., SDOs).
- Between the vehicle and it’s user(s) and other road participants (e.g., other vehicles, bicyclists, and pedestrians).
The understanding and transparency make it easier to understand the limitations of the automated vehicles so that expectations and possibly actions by mentioned actors can be adjusted. From the perspective of use case 4, this means that, for example, it is clear how automated vehicles will respond to reduced speed limits on dynamic road signs because of incidents or traffic jams. Another example can be their response to the information provided via ITS-G5 / cellular about road works. Also, the difference between the desired behaviour and actual behaviour can be input to both road authorities (for road planning and operational measures) and vehicle developers (to improve the vehicle AI).
You can read more about MAPtm at MAPtm | Home