Real-time processing with high classification accuracy is a fundamental requirement in autonomous driving systems. However, existing neural network models for classification often face a tradeoff between computational efficiency and accuracy, necessitating the development of advanced optimization methods to address this limitation. Additionally, dynamic driving environments offer opportunities to enhance classification performance by leveraging the principles of quantum computing, particularly the properties of superposition and entanglement. In response to these challenges, a multi-stage quantum convolutional neural network (MS-QCNN) approach is proposed, designed to improve image analysis performance by effectively utilizing the multi-stage structure of QCNN. A Lyapunov optimization framework is applied to achieve optimal performance, which maximizes time-averaged efficiency while ensuring system stability. This framework dynamically adjusts the MS-QCNN model in response to environmental variations, promoting enhanced queue stability and achieving optimal time-averaged performance.
Real-time processing with high classification accuracy is a fundamental requirement in autonomous driving systems. However, existing neural network models for classification often face a tradeoff between computational efficiency and accuracy, necessitating the development of advanced optimization methods to address this limitation. Additionally, dynamic driving environments offer opportunities to enhance classification performance by leveraging the principles of quantum computing, particularly the properties of superposition and entanglement. In response to these challenges, a multi-stage quantum convolutional neural network (MS-QCNN) approach is proposed, designed to improve image analysis performance by effectively utilizing the multi-stage structure of QCNN. A Lyapunov optimization framework is applied to achieve optimal performance, which maximizes time-averaged efficiency while ensuring system stability. This framework dynamically adjusts the MS-QCNN model in response to environmental variations, promoting enhanced queue stability and achieving optimal time-averaged performance.
This paper presents a novel quantum artificial intelligence model for efficient and accurate pothole detection on road surfaces, specifically tailored for autonomous vehicle (AV) environments. Potholes are widely recognized in the literature as critical risk factors that can cause significant vehicle damage and contribute to traffic accidents. Accordingly, timely and precise detection of potholes is essential to ensure the safety of both vehicles and their occupants. To meet the stringent real-time constraints of AV navigation, the proposed approach employs quantum convolutional neural networks (QCNNs), which provide enhanced expressiveness, low-latency inference, and fewer parameters while achieving performance comparable to classical convolutional neural networks (CNNs). To ensure algorithmic stability under various road and traffic dynamics, a Lyapunov optimization strategy is integrated into the QCNN architecture. This enables the design of a stabilized learning framework, which dynamically regulates the detection model to maintain robustness and performance even in fluctuating environments. Experiments show that the proposed QCNN outperforms classical CNNs in both accuracy and efficiency. In addition, the proposed Lyapunov optimization-based QCNN architecture control further enhances real-time performance, considering inference delay.
This paper proposes a computation-efficient quantum convolutional neural network (CE-QCNN) architecture designed for autonomous driving applications. A key contribution of this work lies in the formulation of a tri-value qubit encoding (TQE) scheme, which compactly embeds three-channel RGB image data into single-qubit states via a sequence of quantum rotations. This strategy enables significant qubit resource reduction while preserving the representational richness of multi-channel visual inputs. The encoded quantum states are subsequently processed through parameterized quantum circuits for convolutional feature extraction, forming the core of the proposed CE-QCNN framework. To further improve learning stability and early-stage performance, a knowledge distillation (KD) strategy is employed, transferring supervision from a pretrained classical CNN model to the quantum network. The proposed model is evaluated on the KITTI dataset, a standard benchmark for autonomous driving, where it demonstrates both competitive detection accuracy and reduced computational complexity. These results substantiate the scalability and practical applicability of CE-QCNNs for future quantum-enhanced perception systems in real-time autonomous driving scenarios.
Autonomous driving systems demand real-time processing of high-dimensional sensor data, large-scale decision making, NP-hard routing optimization under strict latency and computational constraints. Classical approaches using convolutional neural networks face scalability issues and performance limitations in dynamic environments. IBM’s Qiskit framework provides a comprehensive solution with a modular architecture, enabling the construction of quantum circuits that encode multi-sensor data into exponentially large Hilbert spaces. The core components like Terra and Aer support circuit design and noise-aware simulation. Additional libraries extend these capabilities to support quantum machine learning and optimization algorithms. This paper presents a unified Qiskit-based framework incorporating quantum neural networks for trajectory prediction, quantum convolutional neural networks for safety classification in poor conditions, and quantum approximate optimization algorithm for efficient vehicle routing. We showcase how quantum superposition explores multiple routing paths in parallel and how quantum circuits represent grid locations to improve path finding. Our survey reveals measurable performance advantages of quantum-enhanced models over classical methods, forming the basis of quantum-accelerated autonomous driving systems.
Convolutional Neural Networks (CNNs) are extensively utilized for image classification due to their ability to exploit data correlations effectively. However, traditional CNNs encounter several challenges, including limitations in accuracy improvement, computational efficiency, explain ability, and generalization. As data dimensionality or model complexity increases, CNNs often become more difficult to train efficiently, leading to slower processing times. To overcome these challenges, Quantum Convolutional Neural Networks (QCNNs) offer a promising quantum-based approach to enhance existing models, particularly by improving training efficiency and reducing processing times. This paper provides a comparative analysis of classical CNNs and a novel quantum circuit architecture tailored for image-based tasks. The analysis emphasizes the adaptability and potential of quantum circuits to enhance feature extraction, recognition accuracy, and processing efficiency.We applied various QCNN architectures to the Chinese-MNIST dataset and compared their performance with traditional CNN models. Although QCNNs achieved slightly lower accuracy, they demonstrated significantly greater stability. These findings high- light QCNNs’ potential to overcome the computational limitations of conventional deep learning, offering a promising pathway for quantum-enhanced solutions in computer vision. Future work will focus on refining quantum circuit designs, developing hybrid quantum-classical models, and extending QCNN applications to larger and more diverse datasets.
Image classification plays a vital role in large-scale data analysis, especially in object recognition tasks using advanced Deep Learning (DL) frameworks. However, the growing complexity and computational demands of modern DL models have introduced challenges related to scalability and efficiency. Quantum Computing (QC) has emerged as a promising alternative, capable of addressing these limitations by leveraging the principles of Quantum Machine Learning (QML). However, many existing QML models require a large number of qubits, which poses limitations within the current Noisy Intermediate-Scale Quantum (NISQ) era. This work introduces a novel Adaptive Quantum Convolutional Neural Network (AQCNN) designed for efficient and scalable image classification within the constraints of NISQ devices. Addressing the limitations of existing QML approaches, particularly the high qubit requirements, AQCNN incorporates a resource-efficient quantum convolutional layer that performs localized quantum filtering using parameterized quantum circuits. A classical preprocessing layer encodes input features to reduce qubit load, followed by quantum embedding and hybrid quantum-classical layers that optimize feature extraction and classification performance. The model leverages an adaptive quantum convolution strategy, minimizing quantum gate depth and circuit complexity while preserving critical spatial hierarchies in image data. Evaluated on benchmark datasets, AQCNN achieved 95.88% accuracy on MNIST and 95.68% on FMNIST, outperforming comparable QML architectures. Additionally, the model supports scalable execution through parallel quantum circuit arrays, enabling practical deployment on current quantum hardware. This architecture demonstrates a significant advance in quantum-assisted image classification, balancing performance with qubit and gate efficiency. The integration of adaptive quantum convolution and hybrid processing not only enhances classification accuracy but also provides a viable path forward for deploying QML solutions under realistic hardware constraints.
As an emerging research direction in quantum computing, quantum convolutional neural networks provide a promising way to combine quantum feature processing with traditional machine learning algorithms; however, their development is constrained by the scarcity of quantum resources. In this work, we propose a quantum-classical hybrid convolutional neural network model based on quantum convolution kernels. This model employs a multi-data-per-qubit rotation-angle encoding scheme to embed multiple data points into a small number of qubits, aiming to reduce quantum resource consumption and enhance feature extraction for image classification tasks. Its compact structure is designed to extract multidimensional features with fewer qubits and reduced measurement overhead, thereby improving the utilization of limited quantum resources under the considered setting. To evaluate our method, we conducted several experiments to compare our model with existing approaches and to assess performance under varying numbers of qubits, different CNOT gate configurations, and diverse encoding schemes. Experimental results show that the proposed model achieves competitive performance and slightly higher average accuracy than the traditional convolution structure and three existing quantum convolution schemes under a Qiskit Aer noisy simulation environment. This investigation provides a compact design for the structural optimization and low-resource implementation of quantum convolutional neural networks, offering a potential scheme for quantum-assisted image feature extraction.