Camera, LiDAR, and Radar Sensor Fusion Based on Bayesian Neural Network (CLR-BNN)

Ravindran, Ratheesh; Santora, Michael; Jamali, Mohsin M.

doi:10.1109/jsen.2022.3154980

Cited by 31 publications

(6 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Radar-Lidar Fusion [46], integrating radar and Lidar data, can enhance object detection and tracking capabilities, especially in scenarios where accurate distance estimation is critical, while Vision-Lidar Fusion [47], combining vision and Lidar data, can provide detailed information about the shape and characteristics of objects, improving the overall understanding of the scene. Therefore, implementing sensor fusion algorithms involves sophisticated data processing and fusion techniques, such as the Bayesian Neural Network, which combine camera, lidar, and radar sensors [48], or deep learning approaches. These techniques aim to merge information from different sensors while considering their individual strengths and weaknesses, resulting in a more reliable and robust perception system for ACC and other ADAS applications.…”

Section: Vision-based Accmentioning

confidence: 99%

Comparative Analysis of Following Distances in Different Adaptive Cruise Control Systems at Steady Speeds

Mohammed,

Horváth

2024

WEVJ

View full text Add to dashboard Cite

Adaptive Cruise Control (ACC) systems have emerged as a significant advancement in automotive technology, promising safer and more efficient driving experiences. However, the performance of ACC systems can vary significantly depending on their type and underlying algorithms. This research presents a comprehensive comparative analysis of car-following distances in different types of Adaptive Cruise Control systems. We evaluate and compare three distinct categories of ACC systems using three different commercial vehicles brands. The study involves extensive real-world testing at Zalazone Proving Ground, to assess the performance of these systems under various driving conditions, including driving at multiple speeds and applying different car following scenarios. The study investigates how each ACC system manages the minimum following distances according to the type of ACC sensors in each tested vehicle. Our findings revealed that at low to medium ranges of constant driving speeds, there was an approximate linear increase in the average clearances between the two following vehicles for all applied scenarios, with comparatively shorter clearances obtained by the vision-based ACC system, while unstable measurements with a high level of dispersion for all ACC systems were observed at high range of driving speeds.

show abstract

Section: Vision-based Accmentioning

confidence: 99%

Comparative Analysis of Following Distances in Different Adaptive Cruise Control Systems at Steady Speeds

Mohammed,

Horváth

2024

WEVJ

View full text Add to dashboard Cite

show abstract

“…External perception covers information on static and dynamic objects on the street, traffic, and street signs, and is obtained using sensors such as cameras, radars, and LiDARs. External perception provides a real-time picture of the dynamic environment around the vehicle, either by advanced processing of data from a sensor modality [3], [4], [5] or by fusing data from multiple modalities [6], [7]. Another aspect of external perception is localization to determine the location of the vehicle.…”

Section: Generic Adas/ads Architecturementioning

confidence: 99%

Sensing and Machine Learning for Automotive Perception: A Review

et al. 2023

View full text Add to dashboard Cite

Automotive perception involves understanding the external driving environment and the internal state of the vehicle cabin and occupants using sensor data. It is critical to achieving high levels of safety and autonomy in driving. This article provides an overview of different sensor modalities, such as cameras, radars, and light detection and ranging (LiDAR) used commonly for perception, along with the associated data processing techniques. Critical aspects of perception are considered, such as architectures for processing data from single or multiple sensor modalities, sensor data processing algorithms and the role of machine learning techniques, methodologies for validating the performance of perception systems, and safety. The technical challenges for each aspect are analyzed, emphasizing machine learning approaches, given their potential impact on improving perception. Finally, future research opportunities in automotive perception for their wider deployment are outlined.

show abstract

“…LiDAR is a representative vision sensor used in conjunction with cameras and radars. These three sensors have dif-ferent advantages and disadvantages depending on their operating principles and characteristics [18]- [21]. These sensors are used together to complement each other's shortcomings, but they continue to develop to maximize their advantages and complement their shortcomings.…”

Section: Introductionmentioning

confidence: 99%

A Novel Cycloidal Scanning LiDAR Sensor Using Risley Prism and Optical Orthogonal Frequency-Division Multiple Access for Aerial Applications

Kim,

Ashraf,

Eom

et al. 2024

IEEE Access

View full text Add to dashboard Cite

This study proposes a novel light detection and ranging (LiDAR) sensor. The sensor integrates a variable field-of-view (FoV) function by utilizing optical orthogonal frequency-division multiple access (OOFDMA) to generate laser pulse streams and a Risley prism for steering. Using this approach, the LiDAR sensor can adjust the scanning pattern and density of the point cloud to match different sizes, distances, and distributions of objects. To ensure security against external threats, the LiDAR sensor incorporates advanced encryption standards (AES) and wavelength selection techniques. Additionally, the sensor employs various methods to achieve accurate and reliable measurements while minimizing mutual interference between laser signals. This innovative sensor has the potential to enhance the reliability and accuracy of LiDAR technology in aerial applications, thus facilitating the development of more advanced urban air mobility (UAM) sensors for diverse applications. The study evaluated a prototype LiDAR sensor that utilizes a coded laser pulse stream and OOFDMA technology for distance measurement. The LiDAR sensor incorporates three laser diodes with different wavelengths and a Risley prism for steering. The experimental results demonstrate that the proposed LiDAR sensor outperforms existing sensors in terms of accuracy and precision.

show abstract

Camera, LiDAR, and Radar Sensor Fusion Based on Bayesian Neural Network (CLR-BNN)

Cited by 31 publications

References 39 publications

Comparative Analysis of Following Distances in Different Adaptive Cruise Control Systems at Steady Speeds

Comparative Analysis of Following Distances in Different Adaptive Cruise Control Systems at Steady Speeds

Sensing and Machine Learning for Automotive Perception: A Review

A Novel Cycloidal Scanning LiDAR Sensor Using Risley Prism and Optical Orthogonal Frequency-Division Multiple Access for Aerial Applications

Contact Info

Product

Resources

About