Quantum-Cognition-Inspired Decision Making in Autonomous Agents

Vanja Matović

Published Oct 5, 2026• 0 citations

Abstract

Decision-making in autonomous agents requires consideration of physical constraints, safety, uncertainty, learned experience, and mission objectives. Inspired by quantum cognition, this paper presents a hybrid quantum–classical architecture that organizes these aspects into four functions: physical reality, experience, machine conscience, and attention. Physical reality constrains the available actions, experience proposes suitable alternatives, machine conscience evaluates them against predefined safety and operational criteria, and attention selects the final action. Classical checks enforce hard constraints and score soft preferences, while Grover circuits search for actions meeting rule-based thresholds. A conditional Quantum Circuit Born Machine (QCBM) provides learned action preferences. A classical hierarchy gives priority to emergencies, feasibility, risk, and uncertainty, then selects an eligible action using the integration score. The QCBM is trained offline on 2147 examples from a 3060-example teacher-labelled dataset, with the remaining examples reserved for validation and testing. On 466 held-out examples, the QCBM reproduced the teacher-defined preferred action with 83.26% accuracy. In repeated fixed-state replays, all constraint-aware policies had zero hard-constraint violations and full emergency-STOP compliance. The full hierarchy selected STOP less often than the hard-masked QCBM, with the same observed compliance. Finite-sampling experiments showed that higher shot counts reduced probability error and generally brought decisions closer to those obtained from the exact simulated probability distribution. Overall, the results demonstrate that quantum generative learning and Grover-based quantum search can be integrated into a constraint-preserving decision architecture. The results establish functional integration under simulation rather than quantum advantage or physical robot performance. They also demonstrate how quantum technologies can support autonomous decision making and provide a basis for further research in quantum robotics.

DOI: 10.20944/preprints202610.0225.v1

Related papers

Defining Quantum Agents: Formal Foundations, Architectures, and NISQ-Era Prototypes

Eldar Gunter Sultanow, Madjid Tehrani, Siddhant Dutta +2 more

Quantum computing offers potential computational advantages, yet its integration into autonomous decision-making systems remains largely unexplored. This paper addresses the need for a unified framework that systematically combines quantum computation with agent-based artificial intelligence. We examine how quantum technologies can enhance the capabilities of autonomous agents and, conversely, how agentic AI can support the advancement of quantum systems. We analyze both directions of this synergy and present conceptual and technical foundations for future quantum–agentic platforms. Our work introduces a formal definition of quantum agents and outlines architectures that integrate quantum computing with agent-based systems. As concrete proof-of-concept implementations, we develop and evaluate three quantum agent prototypes: (i) a Grover-based decision agent for quantum search-driven action selection, (ii) a variational quantum reinforcement learning agent for adaptive policy learning in a multi-armed bandit setting, and (iii) an adaptive quantum image encryption agent that autonomously selects encryption strategies based on entropy-driven feedback. These prototypes demonstrate practical realizations of quantum agency in decision-making, learning, and security contexts under NISQ-era constraints. Furthermore, we discuss application domains including quantum-enhanced optimization, hybrid quantum–classical orchestration, autonomous quantum workflow management, and secure quantum information processing. By bridging these fields, we introduce a structured theoretical and architectural framework for quantum–agentic systems, providing formal definitions, system models, and early operational prototypes that illustrate the feasibility of quantum-enhanced agency under NISQ constraints.

Quantum ReportsMar 13, 20261 citation

Quantum machine learning for autonomous agents

Meenakshi Munjal, Rajanish Kumar Kaushal

Autonomous agents are being used in more complex and dynamic environments where efficient learning, adaptive decision-making, and scalable optimization are required. Although classical machine learning (ML) and reinforcement methods have greatly contributed to agent intelligence, they are usually restricted by constraints of computational complexity, scaling, and convergence to high-dimensional decision space. The current developments in quantum computing have created new possibilities to solve these problems using quantum ML (QML). In this chapter, the authors discuss the implementation of QML in collaboration with autonomous agents in the larger paradigm of agentic artificial intelligence (AI) and quantum AI agents. It offers a hybrid classical–quantum architectural model where quantum learning modules improve perception, policy learning, and decision optimization, and do not conflict with classical agent pipelines. The chapter analyzes application cases in autonomous robotics, intelligent transportation systems, multi-agent coordination, smart energy management, as well as communication networks, and the chapter shows the potential benefits of quantum-enhanced learning in these areas. The issue of quantum hardware neediness, data encoding, scaling, hybrid integration, and real-life application are critically evaluated and unanswered. In summary, the chapter is a well-organized approach to the purpose of QML in the development of the next-generation autonomous agent and an insight into creating practical, scalable, and intelligent quantum AI agents.

CRC Press eBooksOct 21, 20260 citations

An Architecture for a Quantum Teleo-Reactive Robot

Antonio Chella, Salvatore Gaglio, Giovanni Pilato +1 more

A reactive agent operating in a complex environment must classify its perceived state and select an action under uncertainty. This uncertainty may arise from sensor noise, ambiguous perceptual configurations, or the limited separability of the action regions induced by the agent's policy. We propose a hybrid classical-quantum architecture for a reactive agent in which the perceived state, represented as a classical sensor vector, is mapped onto a quantum feature space. In this space, learned conceptualizations or rule-defined perceptual regions are represented as reference states, and similarities between the current perception and such references are used to support action selection. The architecture is evaluated on a public wall-following robot dataset. Two implementations are considered: (i) a quantum-kernel classifier based on ZZ feature maps and (ii) an illustrative quantum circuit that explicitly encodes sensor conditions into qubits and performs measurement-based action selection. The experimental evaluation is intended as an offline proxy for reactive decision-making, not as a demonstration of a complete closed-loop robotic controller or of quantum advantage. The results show that the proposed framework can represent perceptual ambiguity and connect quantum-state measurement to the selection of discrete reactive actions.

EntropyJun 27, 20260 citations

Hybrid Quantum-AI Architectures for Intelligent Decision Systems

Andreas Novak, Marta Moreau, Isabella Klein

Intelligent decision systems -- AI architectures that synthesise multi-source information to produce actionable decisions in complex environments -- increasingly encounter problem structures where quantum computational resources could provide meaningful advantage: high-dimensional state spaces, combinatorial action selection, uncertainty quantification under exponential hypothesis spaces, and real-time optimisation under resource constraints. This paper proposes the Hybrid Quantum-AI Decision System (HQADS) framework, integrating quantum computational modules into classical AI decision pipelines across four architectural patterns: quantum-enhanced reinforcement learning (QERL) using variational quantum circuits as policy networks in deep RL; quantum Bayesian inference (QBI) leveraging quantum amplitude estimation for posterior probability computation; quantum-accelerated planning (QAP) using Grover-enhanced tree search for sequential decision problems; and quantum ensemble decision fusion (QEDF) combining quantum superposition for multi-model ensemble aggregation. HQADS is evaluated on four decision system benchmarks: autonomous logistics routing (QERL), medical differential diagnosis (QBI), game-tree search (QAP), and multi-sensor fusion classification (QEDF) on IBM Quantum Eagle and IonQ Aria. QERL achieves 18.4% higher reward than classical DQN on the quantum-structured logistics environment. QBI reduces posterior computation time by 6.4x vs. classical MCMC at equal accuracy. QAP achieves 4.8x speedup for depth-12 game-tree search. QEDF improves multi-sensor classification AUC by 3.8pp vs. classical ensemble. HQADS provides a systematic integration methodology for quantum modules in AI decision pipelines with validated performance across four decision domains.

Quantum Frontiers JournalMar 28, 20250 citations

Empowering Autonomous Decision-Making Through Quantum Reinforcement Learning and Cognitive Neuromorphic Frameworks

Balasubramani R, Vidyadevi G. Biradar

The concept of Agentic AI, which focuses on empowering autonomous systems with human-like decision-making capabilities, has gained substantial attention in recent years. By leveraging advanced reinforcement learning (RL) techniques and cognitive frameworks, Agentic AI aims to enable machines to make complex decisions in dynamic environments with minimal human intervention. This paper introduces Agentic AI, a novel approach that combines Quantum Reinforcement Learning (QRL) and Cognitive Neuromorphic Frameworks to empower autonomous decision-making in complex environments. The proposed model leverages quantum computing principles to accelerate learning in reinforcement-based systems, while cognitive neuromorphic frameworks enhance adaptability and real-time decision-making by mimicking human brain functionality. The integration of QRL allows for faster convergence in learning algorithms, improving decision accuracy by 25% and reducing training time by 40%. Cognitive neuromorphic architectures offer enhanced scalability and resilience in dynamic environments, resulting in a 35% improvement in system robustness during testing. Simulations in autonomous navigation and robotic control show that Agentic AI achieves 20% more efficient task execution compared to traditional AI models. These values underscore the potential of the proposed system to revolutionize autonomous systems, providing a robust foundation for decision-making across a range of applications.

International Conference on Mobile Networks and Wireless Communications (ICMNWC)Dec 4, 20244 citations

Survey Paper on Quantum Deep Reinforcement Learning for Robot Navigation Tasks

Shruti Doiphode, Prof. Mukul Jagtap

Abstract: Quantum deep reinforcement learning is an integration of the principles of quantum computing with deep reinforcement learning in an effort to advance robot navigation tasks. Due to the explosion of the state and action spaces, traditionally DRL methods are limited in scalability and efficiency. QDRL, however, uses quantum superposition and entanglement for more efficient processing of big amounts of information-it enables robots to learn optimal policies for navigating complex environments. We introduce here a new QDRL framework in which quantum neural networks encode policy functions and value estimates. We demonstrate how quantum circuits can take advantage of multi-dimensional state representations in order to strengthen the power of a robot in changing its environment over time. Benchmark experiments on navigation tasks reflect phenomenal acceleration in learning performance along with better decision-making accuracy compared to classical DRL methods. We then consider integrating such quantum algorithms, as Grover's search, to improve exploration strategies in sparse reward settings. The experiments show that QDRL speeds up convergence rates and provides robots with the possibility to react more rapidly to unexpected obstacles. It discusses new applications for quantum computing of autonomous systems, potentially applied to robotics, autonomous vehicles, and other environments which require complex decision making. Some lines of work in the future are optimization of the resources needed for quantum computers and hybrid architectures for the resolution of classical problems in navigation

International Journal for Research in Applied Science and Engineering TechnologyOct 18, 20240 citations