Analysis of Automotive Camera Sensor Noise Factors and Impact on Object Detection

Li, Boda; Chan, Pak Hung; Baris, Gabriele; Higgins, Matthew D.; Donzella, Valentina

doi:10.1109/jsen.2022.3211406

Cited by 17 publications

(12 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, comparing point clouds against reference data allows for evaluating sensor data quality, particularly in adverse weather conditions or other degradation events, with respect to AAD system safety requirements. Previous studies have highlighted the significant impact of degradation caused by poor weather conditions on perception algorithms [5], [6], [23]. Additionally, ensuring the safety of autonomous driving systems requires extensive testing, which is impractical through real-world driving alone.…”

Section: A Point Cloud Comparisonmentioning

confidence: 99%

A Novel Score-Based LiDAR Point Cloud Degradation Analysis Method

Shahbeigi,

Robinson,

Donzella

2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

Assisted and automated driving systems critically depend on high-quality sensor data to build accurate situational awareness. A key aspect of maintaining this quality is the ability to quantify the perception sensor degradation through detecting dissimilarities in sensor data. Amongst various perception sensors, LiDAR technology has gained traction, due to a significant reduction of its cost and the benefits of providing a detailed 3D understanding of the environment (point cloud). However, measuring the dissimilarity between LiDAR point clouds, especially in the context of data degradation due to noise factors, has been underexplored in the literature. A comprehensive point cloud dissimilarity score metric is essential for detecting severe sensor degradation, which could lead to hazardous events due to the compromised performance of perception tasks. Additionally, this score metric plays a central role in the use of virtual sensor models, where a thorough validation of sensor models is required for accuracy and reliability. To address this gap, this paper introduces a novel framework that evaluates point clouds dissimilarity based on highlevel geometries. Contrasting with traditional methods like the computationally expensive Hausdorff metric which involves correspondence-search algorithms, our framework uses a tailored downsampling method to ensure efficiency. This is followed by condensing point clouds into shape signatures which results in efficient comparison. In addition to controlled simulations, our framework demonstrated repeatability, robustness, and consistency, in highly noisy real-world scenarios, surpassing traditional methods.INDEX TERMS 3D point cloud, noise factor, perception and sensing, sensor degradation

show abstract

Section: A Point Cloud Comparisonmentioning

confidence: 99%

A Novel Score-Based LiDAR Point Cloud Degradation Analysis Method

Shahbeigi,

Robinson,

Donzella

2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, despite the current advancements of sensor technologies and unique features of each perception sensor [20],degradation of measurements due to sensor noise is inevitable [21]. Sensor noise refers to the random variation inherent in the measurements that can be attributed to a wide range of factors from intrinsic limitations, such as a low resolution or design imperfections, to extrinsic influences, such as atmospheric conditions or electromagnetic interference [22].…”

Section: Introduction 1backgroundmentioning

confidence: 99%

Robust Decision-Making for the Reactive Collision Avoidance of Autonomous Ships against Various Perception Sensor Noise Levels

Lee,

Theotokatos,

Boulougouris

2024

JMSE

View full text Add to dashboard Cite

Autonomous ships are expected to extensively rely on perception sensors for situation awareness and safety during challenging operations, such as reactive collision avoidance. However, sensor noise is inevitable and its impact on end-to-end decision-making has not been addressed yet. This study aims to develop a methodology to enhance the robustness of decision-making for the reactive collision avoidance of autonomous ships against various perception sensor noise levels. A Gaussian-based noisy perception sensor is employed, where its noisy measurements and noise variance are incorporated into the decision-making as observations. A deep reinforcement learning agent is employed, which is trained in different noise variances. Robustness metrics that quantify the robustness of the agent’s decision-making are defined. A case study of a container ship using a LIDAR in a single static obstacle environment is investigated. Simulation results indicate sophisticated decision-making of the trained agent prioritising safety over efficiency when the noise variance is higher by conducting larger evasive manoeuvres. Sensitivity analysis indicates the criticality of the noise variance observation on the agent’s decision-making. Robustness is verified against noise variance up to 132% from its maximum trained value. Robustness is verified only up to 76% when the agent is trained without the noise variance observation with lack of its prior sophisticated decision-making. This study contributes towards the development of autonomous systems that can make safe and robust decisions under uncertainty.

show abstract

“…In other words, some parts of the DNNs are non-deterministic by design. Also, noise and input degradation can affect how well the DNNbased models work in practice [6]. Hence, detecting errors in DNN-based systems fundamentally differs from traditional automation systems (e.g., anti-lock braking system) and is hereafter defined with a different term, "introspection".…”

Section: Introductionmentioning

confidence: 99%