Symbolic Logic Framework for Situational Awareness in Mixed Autonomy

Period: Oct 2022 – Sept 2024
Role: Project Leader
Principal Investigator: Dr. Sofie Haesaert
Affiliation: Eindhoven University of Technology (TU/e)
Funding: European Innovation Council (EIC) – Horizon Europe
Consortium: TU/e, MPI-SWS, KTH, Uppsala University, Siemens, Netherlands Aerospace Centre (NLR)
Website: https://www.symaware.eu/

Overview

SymAware develops a symbolic and logical framework for situational awareness in multi-agent systems (MAS) operating in mixed human–autonomy environments.

Modern autonomous systems increasingly interact with humans and other autonomous agents, yet current architectures lack formal mechanisms for representing and reasoning about knowledge, uncertainty, risk, and social interaction.

The goal of SymAware is to establish a formal foundation for awareness-aware autonomy, enabling intelligent agents to reason about what they know, what they do not know, and how risks evolve in dynamic multi-agent environments.

Technical Focus

The project investigates how symbolic reasoning and formal methods can support structured situational awareness across multiple dimensions:

• Logical characterization of awareness: Formalizing awareness using compositional and symbolic logic.

• Spatiotemporal reasoning for decision-making: Integrating spatial and temporal reasoning into awareness models.

• Risk-aware reasoning under uncertainty: Quantifying risks when agents possess incomplete or uncertain knowledge.

• Knowledge awareness and communication: Enabling agents to reason about the knowledge and awareness of others.

• Human–autonomy collaboration: Supporting safe cooperation between humans and autonomous agents.

My Contribution

As project leader and postdoctoral researcher, I led research on formal reasoning and risk-aware decision-making for autonomous systems within the SymAware framework.

My work focused on bridging formal logic with control and decision-making, including:

• Designing risk-aware reasoning mechanisms that enable agents to reason under incomplete or uncertain knowledge

• Investigating how formal awareness models can inform safe decision-making and control strategies

• Contributing to the integration of symbolic reasoning with uncertainty quantification in autonomous systems

These efforts contribute toward a formal framework for risk-aware autonomy, enabling intelligent agents to reason about safety, uncertainty, and interaction in mixed human–machine environments.

Applications and Impact

SymAware investigates how formal awareness models can support safer autonomy in complex environments where multiple agents—both human and autonomous—interact.

The framework is explored in safety-critical domains such as:

• Autonomous driving

• Aviation systems

By integrating symbolic reasoning, uncertainty quantification, and multi-agent interaction models, the project advances a new paradigm of awareness-aware autonomous systems that can explicitly reason about risk, knowledge, and interaction.

This work contributes to the broader goal of trustworthy and explainable autonomy, aligning with the “Awareness Inside” challenge of the European Innovation Council and supporting the safe deployment of autonomous technologies in real-world human–machine ecosystems.

Publications

• S. Qi, Z. Zhang*, S. Haesaert, and Z. Sun, "Automated Formation Control Synthesis from Temporal Logic Specifications",  62nd IEEE Conference on Decision and Control (CDC 2023), Singapore, 13-15 Dec 2023.

• Z. Zhang* and S. Haesaert, "Modularized Control Synthesis for Complex Signal Temporal Logic Specifications",   62nd IEEE Conference on Decision and Control (CDC 2023), Singapore, 13-15 Dec 2023.

• L. C. Wu, Z. Zhang*, S. Haesaert, Z. Ma, and Z. Sun, "Risk-Aware Reward Shaping of Reinforcement Learning Agents for Autonomous Driving", 49th Annual Conference of the IEEE Industrial Electronics Society (IECON 2023), Singapore, 16-19 Oct 2023. 

• N. Dang, T. Shi, Z. Zhang, W. Jin, M. Leibold, and M.Buss, "Identifying Reaction-Aware Driving Styles of Stochastic Model Predictive Controlled Vehicles by Inverse Reinforcement Learning", 26th IEEE International Conference on Intelligent Transportation Systems (ITSC 2023), Bilbao, Spain, 24-28 Sept 2023. 

• Z. Zhang*, Z. Sun, and S. Haesaert, "Risk-Aware Task Assignment with Formal Specifications for Heterogeneous Multi-Agent Systems", 42nd Benelux Meeting on Systems and Control, Elspeet, The Netherlands, 21-23 Mar 2023.