Situation-Aware Interactive MPC Switching for Autonomous Driving

2025-12-05

AI Summary

Model Predictive Control (MPC) strategies for autonomous driving in interactive traffic scenarios often struggle to balance control performance and real-time computational efficiency. Method: This paper proposes a context-aware dynamic MPC controller switching framework. A lightweight neural network-based context classifier online evaluates traffic interaction intensity; accordingly, a simplified MPC is deployed under low-interaction conditions, while a high-fidelity interactive MPC—capable of modeling strategic maneuvers such as game-theoretic lane changes and emergency evasive actions—is automatically activated under high-interaction conditions. Contribution/Results: To our knowledge, this is the first work that jointly designs context classification and multi-level interactive MPC modeling. The framework achieves a Pareto-optimal trade-off between computational cost and safety performance. Experiments demonstrate that the method retains 99.2% of the original high-fidelity MPC’s control quality while reducing average computation time by 63.5%, significantly enhancing system real-time capability and robustness.

Technology Category

• Multiagent Systems: Adversarial Agents
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Mixed Discrete/Continuous Planning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

To enable autonomous driving in interactive traffic scenarios, various model predictive control (MPC) formulations have been proposed, each employing different interaction models. While higher-fidelity models enable more intelligent behavior, they incur increased computational cost. Since strong interactions are relatively infrequent in traffic, a practical strategy for balancing performance and computational overhead is to invoke an appropriate controller based on situational demands. To achieve this approach, we first conduct a comparative study to assess and hierarchize the interactive capabilities of different MPC formulations. Furthermore, we develop a neural network-based classifier to enable situation-aware switching among controllers with different levels of interactive capability. We demonstrate that this situation-aware switching can both substantially improve overall performance by activating the most advanced interactive MPC in rare but critical situations, and significantly reduce computational load by using a basic MPC in the majority of scenarios.

Problem

• Balancing autonomous driving performance and computational cost
• Switching controllers based on traffic interaction demands
• Improving performance in critical situations while reducing load

Innovation

• Hierarchical MPC formulations based on interactive capabilities
• Neural network classifier for situation-aware controller switching
• Balances performance and computational load via adaptive control