International Conference on Computer and Knowledge Engineering

Home / 15th International Conference on Computer and Knowledge Engineering

Autonomous Drone Navigation Using Synchronized Camera and IMU Data with CNN

Authors :

Reza Javanmard Alitappeh¹ Narges Hamzeh Mermeti² Fatemeh Barzegar³ Fatemeh Ebrahimi⁴ Nima Mahmoudi⁵ Jalal Alipour Langouri⁶

1- University of Science and Technology of Mazandaran 2- University of Science and Technology of Mazandaran 3- University of Science and Technology of Mazandaran 4- University of Science and Technology of Mazandaran 5- University of Science and Technology of Mazandaran 6- University of Science and Technology of Mazandaran

Keywords :

Autonomous Drone Navigation،Deep Learning،Convolutional Neural Network (CNN)،Inertial Measurement Unit (IMU)

Abstract :

Accurate motion estimation plays a crucial role in enabling reliable autonomous drone navigation, particularly in GPS-denied environments such as indoor and underground settings. This paper proposes a deep learning–based framework that predicts the drone’s dynamic behavior using synchronized visual and inertial data. The approach leverages monocular camera frames aligned with inertial measurement unit (IMU) signals to train a Convolutional Neural Network (CNN) for estimating both linear and angular accelerations. The data collection process employs ROS bag files to ensure precise temporal alignment and consistent input-output mapping. The network architecture is optimized to extract spatial-temporal features from the sensor stream while maintaining computational efficiency for real-time performance. Without relying on GPS or external localization systems, the proposed method provides an effective and scalable solution for robust motion prediction and stable drone control. Experimental results show consistent generalization across varying trajectories, confirming the model’s reliability in complex, constrained environments.‌