Autonomous Driving Control Benchmark Challenge
Autonomy and electrification stand as pivotal trends in the automotive industry. Motion control, particularly for electrified vehicles, emerges as a cornerstone technology essential for implementing autonomous driving. However, within the academic realm of control and decision-making, this critical issue has not garnered significant attention, despite being addressed within the vehicular engineering field over the past few decades.
The primary objective of this benchmark problem is to establish a platform for students and emerging researchers to confront the challenges inherent in next-generation autonomous driving control. Additionally, it aims to facilitate the exchange of cutting-edge research findings in automotive system control and optimization. The benchmark problem presents a straightforward yet profound motion control formulation tailored for four in-wheel motor-actuated electric vehicles, drawing inspiration from the core technology of autonomous driving.
These new themes have enabled University of L'Aquila to take into accoutn the realities of the field by recently setting up a new test bench based on fleet scale at 1/10.
Comprised of industrial researchers and academic experts in control, the organizing team endeavors to foster an environment conducive to innovation and collaboration.
The laboratory has some systems such as:.
a) 1-10 scale connected and autonomous electric vehicles, for navigation trajectory planning purposes in environments with uncertainties.
b) A simulator of hybrid electric vehicles aimed at evaluating performance from an energy point of view for designing control strategies to minimize consumption and emissions.
Open Topics for Master's Theses
- Machine learning approach in control loop for scaled 1/10 connected and autonomous cars - The research explores the applications of autonomous vision and decision-making across various fields, particularly in driverless and connected cars and highway decision-making. It emphasizes the significant role of machine learning in providing robust and adaptive performance while also reducing development costs and time. The study will introduce features such as detection capability, applicable terrain, and different detection radar technologies. Thesis Proposal E-Pico_1
- Robust Intersection Management for Connected Autonomous Vehicles - This thesis delves into the safety and performance challenges associated with implementing intersession of autonomous vehicles. It introduces a time and space aware technique designed to address these challenges by being robust against model mismatches, external disturbances, and nondeterministic delays in network and processing time. Thesis Proposal E-Pico_2
- Control Architecture for Connected Vehicle Platoons - From sensor data to controller design using vehicle-to-everithing communication - This thesis proposal deals with robust cooperative control of connected and automated vehicles (CAVs) with special consideration of the coupled safety inter-vehicle distance constraints, under stochastic communication delays, state noises, and channel interferences. To address the hard constraints caused by the vehicle physical limitations and the coupled safety constraints among adjacent vehicles explicitly, the distributed model predictive control (DMPC) problem for CAVs is formulated. Thesis Proposal E-Pico_3
- Autonomous Freight Transportation and Drone Delivery - The logistics and delivery industry is facing challenges such as high transportation costs, difficulty in meeting customer demands, and environmental concerns. However, the integration of drone and autonomous vehicle technology can address these challenges by reducing transportation costs, increasing speed and reliability of delivery, and improving efficiency. Thesis Proposal E-Pico_4
- Research on the collaboration and collision avoidance of multiple automated vehicles - Research on the collaboration and collision avoidance of multiple automated vehicles is being conducted, along with the development of a conceptual framework empowered by AI. This framework aims to enhance communication and decision-making among vehicles to prevent accidents. Thesis Proposal E-Pico_5
- Traffic control for autonomous driving that incorporates human decision-making into the loop - The concept of autonomous driving traffic control based on human-in-the-loop decisions involves integrating human decision-making with autonomous driving systems to regulate traffic flow. Thesis Proposal E-Pico_6
- Dynamic Path-Planning and Charging Optimization for Autonomous Electric Vehicles in Transportation Networks - The thesis proposal introduces a method for dynamic path-planning and charging optimization tailored for autonomous electric vehicles (AEVs) within transportation networks. This approach aims to enhance the efficiency and sustainability of AEV operations by optimizing both route selection and charging strategies. Thesis Proposal E-Pico_7
- Physics-Informed Neural Nets-based Model Predictive Control for Connected Autonomous Vehicles - The thesis proposal introduces a method for control of connected and autonomous vehicles using physic-informed neural network in particular situations. The absence of data or computational complexity are aspects to take into consideration. Thesis Proposal E-Pico_8
Completed for Master's Theses
1) "Control of Automated Driving in Motion Planning", Yucheng Li, Master Science Degree (2022), Advisor: Stefano Di Gennaro.
2) "Design of Vision-based Control Architectures towards Autonomous Vehicles", Chiara Romano, Master Science Degree (2023), Advisor: Stefano Di Gennaro, Correlator: Giovanni De Gasperis.
3) "Analysis of electrical safety requirements for electric vehicles and implementation of electrical risk safety measures", Jimena Jimenez Hernandez, Master Science Degree (2022), Advisor: Stefano Di Gennaro.
4) "Accident avoidance strategies for vehicles", David Antonio Martìnez Carrillo, Master Science Degree (Co–tutored Cinvestav–Guadalajara, 2017), Advisor: Stefano Di Gennaro.
5) "Active attitude control of ground vehicles with wireless smart tire sensors and performance evaluation in CarSim", Master Science Degree, Andrea Franceschini, Advisor: Stefano Di Gennaro.
The laboratory's educational and scientific activity is linked to the E-PICO Project. This master's programme has been selected under the European Programmes of excellence: in 2019 through the Erasmus Mundus Programme (EPICO Master).
Project Reference: 610569-EPP-1-2019-1-FR-EPPKA1-JMD-MOB (EU Portal Project profile)
Page subject to continuous updating.
Last update: 17-04-2024