DESCRIPTION

This workshop addresses one of the timely topics in the field of autonomous multi-vehicle systems for Smart Cities and Industry 5.0: formation control and collision avoidance for multi-agent configurations. As it is well-known, such issues have received increasing attention, as testified by the relevant number of contributions, see, e.g., [R1]–[R3] and references therein. Essentially, this is due to the success of using Unmanned Vehicles (UVs) in different areas of Industry 5.0 and Intelligent Transportation Systems [R4]. In this context, it is important to underline that the design of a fully autonomous vehicle system must take into account the following key requirements: (i) each vehicle must be able to track a reference signal and reach the desired goal, (ii) formation control tools are needed to guarantee coordination among vehicles, and (iii) absence of collisions must be ensured.

Despite the significant interest, most of the existing approaches provide efficient solutions only when physical limitations are not taken into account in the vehicles’ modelling. As a consequence, their effectiveness and absence of collision claims are not guaranteed in real scenarios where the vehicle’s geometrical constraints and saturation effects play a crucial role.

According to these premises, the workshop is devoted to analyzing recent advances in constrained control of multi-vehicle systems. We will collect new trends and solutions addressing the collision-free reference tracking control problem for constrained multi-vehicle systems moving in shared environments (e.g. Industrial Plants and/or Cities) where obstacles might be present. The distinctiveness of this workshop relies on showing how both traditional constrained control techniques and data-driven approaches can be used standalone/combined to derive novel and more efficient hybrid control architectures for multi-vehicle autonomous systems.

RELEVANCE OF THE PROPOSED TUTORIAL

This workshop is aligned with one of the main conference topics: the tight integration of autonomous vehicles within the horizons of Industry 5.0.  The workshop is dedicated to addressing critical issues in the domain of autonomous multi-vehicle systems, particularly concerning formation control and collision avoidance for configurations involving multiple agents. With the increasing prominence of Unmanned Vehicles (UVs) in various sectors of Industry 5.0 and Intelligent Transportation Systems, the workshop underscores the necessity for fully autonomous vehicle systems to meet key requirements, including precise signal tracking, coordinated formation control, and collision prevention. Despite significant interest, existing approaches often fall short in accounting for physical limitations in vehicle modeling, leading to uncertainties regarding their effectiveness and collision avoidance capabilities. This workshop explores recent advancements in constrained control techniques for multi-vehicle systems, seeking innovative solutions to the challenge of collision-free reference tracking control, especially in shared environments like Industrial Plants and Cities where obstacles may be present. The distinctive aspect of the workshop lies in its endeavour to demonstrate how both traditional constrained control methods and data-driven approaches can be leveraged independently or in combination to develop novel and more efficient hybrid control architectures for autonomous multi-vehicle systems.

WORKSHOP SUPPORT

• IEEE SMC Italian Chapter  
• IFAC Technical Committee 6.4. Fault Detection, Supervision & Safety of Technical Processes-SAFEPROCESS

WORKSHOP ORGANIZATION, TIMELINE, PRESENTATIONS 

This one-day workshop will consist of 45-min lectures provided by invited speakers. To encourage discussions, a Q&A period of 10/15 minutes will be available at the end of each part. Interaction between speakers and attendees will be strongly encouraged also during the talks.

TENTATIVE SCHEDULE

• Part 1 - Talks, Speakers and Abstracts: 
  • First Talk: Fleet Management of Autonomous Mobile Robots for Warehouse Logistics
    • Speaker: Rok Vrabič, University of Ljubljana, Slovenia
    • Abstract: Fleet management in intralogistics with Automated Guided Vehicles (AGVs) and Autonomous Mobile Robots (AMRs) involves interconnected challenges in path planning, dispatching, fleet sizing, facility layout, and roadmap design. This talk highlights the interdependencies of these elements and presents recent advancements in three key areas. First, it explores methods for generating and evaluating roadmaps, crucial for system performance and adaptability. Second, it addresses practical considerations in AMR system implementation. Finally, the presentation focuses on applying reinforcement learning to vehicle dispatching, demonstrating how this data-driven approach can adapt to dynamic intralogistics environments and potentially surpass traditional rule-based methods.
  • Second Talk: Safe Coordination of Autonomous Vehicles
    • Speaker: Vicenç Puig , Universitat Politècnica de Catalunya, Spain
    • Abstract: This talk will present some recent results regarding the safe coordination of autonomous vehicles. In particular, the talk will show how by means of the LPV approach the modelling of autonomous vehicles (including both kinematic and dynamic equations) can be addressed both from a physical perspective and by using data-based learning techniques. Then, using the LPV representation of the vehicle, efficient planning and control implementation based on MPC and LMIs will be presented. The inclusion of uncertainty in the modelling using sets (zonotopes) will also be illustrated. Finally, the extension of the proposed approaches to deal with learning-based algorithms that allow to estimate the vehicle model (dynamic part) from data is presented. The algorithms will  be illustrated using high-fidelity simulators and small scale/real size vehicles.

• Break 

• Part 2  - Talks, Speakers and Abstracts:  
  • Third Talk: Application of Deep Reinforcement Learning for Traffic Control of Road Intersection with Autonomous Vehicles 
    • Speaker: Agostino Marcello Mangini, Polytechnic of Bari, Italy
    • Abstract:  The control and management of road intersection in presence of Autonomous Vehicles (AVs) is central in terms of safety requirements for Cooperative, Connected and Automated Mobility (CCAM). In this talk a control methodology based on Deep Reinforcement Learning is applied both at the traffic light and at unsignalized intersections. The traffic light system is controlled by the DRL agent, which can get a “snapshot” of the state of the intersection at each agent-step, observing the oncoming vehicles on each arm, and their respective priorities. Observing the road in a limited area near the traffic light, the agent uses this information to properly select one of the fixed sets of traffic light phases. The aim is efficiently managing the traffic in order to minimize the global waiting time of vehicles at the intersections.
      On the other hand, the intersections can be unsignalized. In this context, a novel approach to optimize traffic flow at unsignalized intersections by using cooperative deep reinforcement learning is discussed in this talk. In the proposed multiagent strategy, each connected automated vehicle (CAV) acts as an agent and has partial observability of the current state made of other CAVs, regular vehicles and priority connected vehicles (CPV), such as ambulances and police cars. The goal of each agent is to give right-of-way to priority vehicles, prevent collisions and optimize intersection crossing time.
  • Forth Talk: Enhanced Traffic Modeling and Control for Freeway Networks in the Era of Connected and Automated Vehicles
    • Speaker: Silvia Siri,  University of Genova, Italy
    • Abstract: For decades, researchers have studied traffic models and explored control techniques for road traffic networks with the primary goal of reducing congestion and enhancing the citizens' quality of life. However, in recent years, the advancements in measurement, communication, and computing technologies have revitalized research in traffic modelling and management. These developments aim to revise conventional models and control algorithms to fully exploit the potential of new vehicle technologies, in particular automated and connected vehicles that will coexist with traditional human-driven vehicles in the roads for the next decades. A significant shift in traffic management is occurring as research transitions from road-based traffic control strategies, relying on fixed actuators and sensors, to vehicle-based control strategies leveraging advanced, increasingly connected, and automated vehicle technologies. This talk focuses on the development of new traffic models for representing the mixed scenario of traditional and automated vehicles sharing the same road, as well as on traffic control for freeway networks based on the communication and automation advancements offered by new vehicular technologies. Among these control methods, the use of platoons of vehicles for the definition of mainstream control strategies is specifically addressed in the talk.

• Break

• Part 3 - Talks, Speakers and Abstracts:   
  • Fifth Talk: Factory Digital Twinning for Autonomous Transportation and Material Handling
    • Speaker: Hui Yang, The Pennsylvania State University, Usa
    • Abstract: The new wave of Industry 4.0 is transforming manufacturing factories into data-rich environments. This provides an unprecedented opportunity to feed large amounts of sensing data collected from the physical factory into the construction of digital twin (DT) in cyberspace. There is an urgent need to exploit the full potential of DT technology and simulation optimization to improve the smartness and autonomous levels of manufacturing transportation and material handling systems. This talk will present the design of new DT models for simulation optimization of manufacturing process flows. The DT will describe nonlinear and stochastic dynamics among a network of interactive manufacturing things, including customers, machines, automated guided vehicles (AGVs), queues, and jobs. Further, new optimization algorithms will be introduced to design sequential computer experiments and optimize the utilization of AGV under uncertainty. We will show the new designs of two new graph models—job flow graph and AGV traveling graph—to track and monitor the real-time performance of manufacturing jobshops. The next generation of digital twin is strongly promised to innovate autonomous transportation and material handling for smart manufacturing. At the end of this talk, future research directions will be discussed.
  • Sixth Talk: Autonomous Vehicle Platoons in Urban Road Networks
    • Speaker: Francesco Giannini, University of Calabria, Italy
    • Abstract: The talk is devoted to disseminating recent advances in constrained control of multi-vehicle systems. We will discuss recent solutions addressing the collision-free reference tracking control problem for constrained multi-vehicle systems moving in shared environments (e.g. Industrial Plants and/or Cities) where obstacles might be present. The distinctiveness of this tutorial relies on showing that traditional Model Predictive Control (MPC) solutions can be combined with recent-developed Reinforcement Learning tools to derive novel and more efficient hybrid control architectures for multi-vehicle autonomous systems.

REFERENCES

[R1] K.-K. Oh, M.-C. Park, and H.-S. Ahn, “A survey of multi-agent formation control,” Automatica, 53: 424-440, 2015.

[R2] J. Dahl, G. R. de Campos, C. Olsson, J. Fredriksson, “Collision avoidance: A literature review on threat-assessment techniques’,’ IEEE Transactions on Intelligent Vehicles 4(1): 101-113, 2018.

[R3] B. Paden, M. Čáp, S. Z. Yong, D. Yershov, E. Frazzoli, “A survey of motion planning and control techniques for self-driving urban vehicles,” IEEE Transactions on intelligent vehicles 1(1):33-55, 2016.

[R4] P. K. R. Maddikunta et al. "Industry 5.0: A survey on enabling technologies and potential applications." Journal of Industrial Information Integration 26: 100257, 2022.