arXiv subjects

Marc-Philippe Neumann

Publications and source records attributed to Marc-Philippe Neumann.

5 recordsLinked to original sources

arXiv2026

Learning-based Multi-agent Race Strategies in Formula 1

In Formula 1, race strategies are adapted according to evolving race conditions and competitors' actions. This paper proposes a reinforcement learning approach for multi-agent race strategy optimization. Agents learn to balance energy management, tire degradation, aerodynamic interaction, and pit-stop decisions. Building on a pre-trained single-agent policy, we introduce an interaction module that accounts for the behavior of competitors. The combination of the interaction module and a self-play training scheme generates competitive policies, and agents are ranked based on their relative performance. Results show that the agents adapt pit timing, tire selection, and energy allocation in response to opponents, achieving robust and consistent race performance. Because the framework relies only on information available during real races, it can support race strategists' decisions before and during races.

cs.AI↗

arXiv2025

Towards Learning-Based Formula 1 Race Strategies

This paper presents two complementary frameworks to optimize Formula 1 race strategies, jointly accounting for energy allocation, tire wear and pit stop timing. First, the race scenario is modeled using lap time maps and a dynamic tire wear model capturing the main trade-offs arising during a race. Then, we solve the problem by means of a mixed-integer nonlinear program that handles the integer nature of the pit stop decisions. The same race scenario is embedded into a reinforcement learning environment, on which an agent is trained. Providing fast inference at runtime, this method is suited to improve human decision-making during real races. The learned policy's suboptimality is assessed with respect to the optimal solution, both in a nominal scenario and with an unforeseen disturbance. In both cases, the agent achieves approximately 5s of suboptimality on 1.5h of race time, mainly attributable to the different energy allocation strategy. This work lays the foundations for learning-based race strategies and provides a benchmark for future developments.

eess.SY↗

arXiv2025

Game Theory in Formula 1: From Physical to Strategic Interactions

This paper presents an optimization framework to model multi-agent racing dynamics. By incorporating physically accurate interaction models and accounting for the optimal responses of competing agents, our approach reveals strategic behaviors typical of motorsport. Aerodynamic wake effects, trajectory optimization, and energy management are captured and evaluated on a representative case study, based on a Formula 1 scenario. We describe the minimum lap time problem with two agents as either a Nash or a Stackelberg game, and by employing the Karush-Kuhn-Tucker conditions during the problem formulation, we recover the structure of a nonlinear program. In addition, we introduce an algorithm to refine local Stackelberg solutions, using the Nash costs as upper bounds. The resulting strategies are analyzed through case studies. We examine the impact of slipstreaming on trajectory selection in corners, straights, and high-speed sections, while also identifying optimal overtaking locations based on energy allocation strategies. Exploiting the structural similarities of the game formulations, we are able to compare symmetric and hierarchical strategies to analyze competitive racing dynamics. The proposed methodology closes the gap between theoretical game theory and practical applications, with relevance in multi-agent systems with coupled nonlinear dynamics.

eess.SY↗

arXiv2024

Game-theoretic Energy Management Strategies With Interacting Agents in Formula 1

This paper presents an interaction-aware energy management optimization framework for Formula 1 racing. The considered scenario involves two agents and a drag reduction model. Strategic interactions between the agents are captured by a Stackelberg game formulated as a bilevel program. To address the computational challenges associated with bilevel optimization, the problem is reformulated as a single-level nonlinear program employing the Karush-Kuhn-Tucker conditions. The proposed framework contributes towards the development of new energy management and allocation strategies, caused by the presence of another agent. For instance, it provides valuable insights on how to redistribute the energy in order to optimally exploit the wake effect, showcasing a notable difference with the behavior studied in previous works. Robust energy allocations can be identified to reduce the lap time loss associated with unexpected choices of the other agent. It allows to recognize the boundary conditions for the interaction to become relevant, impacting the system's behavior, and to assess if overtaking is possible and beneficial. Overall, the framework provides a comprehensive approach for a two-agent Formula 1 racing problem with strategic interactions, offering physically intuitive and practical results.

eess.SY↗

arXiv2023

Low-level Online Control of the Formula 1 Power Unit with Feedforward Cylinder Deactivation

Since 2014, the F\'ed\'eration Internationale de l'Automobile has prescribed a parallel hybrid powertrain for the Formula 1 race cars. The complex low-level interactions between the thermal and the electrical part represent a non-trivial and challenging system to be controlled online. We present a novel controller architecture composed of a supervisory controller for the energy management, a feedforward cylinder deactivation controller, and a track region-dependent low-level nonlinear model predictive controller to optimize the engine actuators. Except for the nonlinear model predictive controller, the proposed controller subsystems are computationally inexpensive and are real time capable. The framework is tested and validated in a simulation environment for several realistic scenarios disturbed by driver actions or grip conditions on the track. In particular, we analyze how the control architecture deals with an unexpected gearshift trajectory during an acceleration phase. Further, we demonstrate how an increased maximum velocity trajectory impacts the online low-level controller. Our results show a suboptimality over an entire lap with respect to the benchmark solution of 49 ms and 64 ms, respectively, which we deem acceptable. Compared to the same control architecture with full knowledge of the disturbances, the suboptimality amounted to only 2 ms and 17 ms. For all case studies we show that the cylinder deactivation capability decreases the suboptimality by 7 to 8 ms.

eess.SY↗