Amirhesam Aghanouri

Novel Certada, Amirhesam Aghanouri, Joseba Gorospe et al.

This paper investigates a multi-modal and human-centric framework for low-cost road roughness assessment. The evaluation was based on three complementary data sources: smartphone-based International Roughness Index (IRI) estimates from two independent smartphone-based applications; in-vehicle GNSS-IMU Receiver (Global Navigation Satellite System Receiver with Inertial Measurement Unit) measurements, and passenger Present Serviceability Ratings (PSR). Data were collected over 1700 km across Austria, Hungary, and Romania under real traffic conditions. Inter-application agreement was evaluated using correlation analysis, Intraclass Correlation Coefficient (ICC), and Bland-Altman methods. While the two smartphone applications show strong correlation, systematic bias limits their interchangeability. A significant inverse relationship between IRI and PSR confirms perceptual sensitivity to roughness, and positive correlations between IRI and vertical acceleration validate the physical linkage between pavement irregularities and vehicle dynamics. The results demonstrate the challenges of integrating consumer-grade sensing and perception-based evaluation for road roughness monitoring as an alternative to high-cost specialized survey equipment.

Amirhesam Aghanouri, Cristina Olaverri-Monreal

Autonomous vehicles (AVs) are expected to revolutionize transportation by improving efficiency and safety. Their success relies on 3D vision systems that effectively sense the environment and detect traffic agents. Among sensors AVs use to create a comprehensive view of surroundings, LiDAR provides high-resolution depth data enabling accurate object detection, safe navigation, and collision avoidance. However, collecting real-world LiDAR data is time-consuming and often affected by noise and sparsity due to adverse weather or sensor limitations. This work applies a denoising diffusion probabilistic model (DDPM), enhanced with novel noise scheduling and time-step embedding techniques to generate high-quality synthetic data for augmentation, thereby improving performance across a range of computer vision tasks, particularly in AV perception. These modifications impact the denoising process and the model's temporal awareness, allowing it to produce more realistic point clouds based on the projection. The proposed method was extensively evaluated under various configurations using the IAMCV and KITTI-360 datasets, with four performance metrics compared against state-of-the-art (SOTA) methods. The results demonstrate the model's superior performance over most existing baselines and its effectiveness in mitigating the effects of noisy and sparse LiDAR data, producing diverse point clouds with rich spatial relationships and structural detail.