MOVECIT-NavEn
Navigation and Energy Optimization for Urban Vehicles
PID2024-157191OB-C21
Proyectos de Generación de Conocimiento 2024
An autonomous vehicle (AV) is a vehicle that integrates not only multi-sensor perception, but also intelligent decision making and control technology. There is no doubt that the use of AVs in a massive way will be a reality worldwide in the coming decades. To achieve this goal, the underlying technologies have to reach a degree of maturity that they have not yet reached. In this context, the automotive sector has adopted a great number of Artificial Intelligence (AI) innovations. The urban environment understanding is key point in this process, where the navigation system must plan its trajectories in a way that does not cause a risk of collision in the current state and also in future states. So, the first aim of the subproject 1 is the urban understanding by developing autonomous navigation algorithms that use AI to interpret the urban environment and optimize route planning in real time, adapting to complex urban environments. Moreover, in this scenario appears the pivotal question of detecting when some unexpected behaviour is occurring, this is referred to as anomaly detection in this project. Anomaly detection or the identification of abnormal instances of data represents a relevant issue across different research topics, where in this project, it has been focused on guaranteeing safety and reliability in monitoring, navigation and energy management.
Secondly, the urban transportation electrification is a key strategy for reducing greenhouse gas emissions and minimizing the carbon footprint of cities. Current urban mobility challenges involve improving sustainability and implementing energy-efficient and emission-control solutions. However, integrating electric vehicles (EVs) into smart grids presents significant challenges. On the one hand, optimizing charging locations for EVs when they are on-route is critical because some charging stations can be overloaded. On the other hand, energy requirement from vehicle fleet depots requires balancing individual vehicle needs with overall installation energy demand at the facility. Although commercial applications for EV fleet management exist at the logistical level, they do not address charging optimization nor evaluate its impact on the power grid, such as line overloads, transformer station stress, or voltage surges. Additionally, EVs can be leveraged as energy service providers through vehicle-to-grid (V2G) technology, offering potential economic benefits to fleet managers.
The subproject 1 is a fundamental part of the proposed multidisciplinary approach of this project that focus on the development of an advanced technological ecosystem for sustainable and intelligent urban mobility by integrating in-cabin monitoring systems, multimodal language models applied to navigation, energy optimization, and fleet management through digital twins, artificial intelligence, and specialized databases. The main goal is to ensure safety, operational efficiency, and scalability of the solutions within the context of Smart Cities. The specific objectives of the subproject 1 into the proposed multidisciplinary approach related to the advanced technological ecosystem for sustainable and intelligent urban mobility are: AI-based multimodal language models applied to navigation, Intelligent Energy Management using AI, Intelligent Fleet Management, Specialized Databases for AI Training, Anomaly Detection through AI and Experimental Validation.
Proyecto PID2024-157191OB-C21 financiado por:
