Multi-sensor Fusion Module in a Fault Tolerant Perception System for Autonomous Vehicles

Realpe, Miguel

doi:10.18178/joace.4.6.460-466

Cited by 8 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al 2 used an artificial intelligence methodology to carry out fault diagnosis for rotating machinery after broadly reviewing literature relevant to industrial applications. Realpe et al 3 used multi-sensor fusion architecture and the support vector machine algorithm to minimize the influence of sensor faults in an autonomous vehicle. Janssens et al 4 applied a convolutional neural network to the fault detection of rotating machinery using vibration analysis.…”

Section: Introductionmentioning

confidence: 99%

Dual-sliding mode approach for separated fault detection and tolerant control for functional safety of longitudinal autonomous driving

Song

Min

et al. 2020

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

This paper describes model-based separated fault detection and fault tolerant control of longitudinal autonomous driving using dual-sliding mode observer for functional safety. Internal and environment sensors such as camera or radar are required to measure the acceleration information of the subject vehicle and the relative distance and velocity information between the preceding and subject vehicles in longitudinal autonomous driving. In order to detect the independent fault of each sensor, a dual-sliding mode observer (SMO) is used for fault reconstruction under the assumption that V2V (Vehicle to Vehicle) communication for vehicle driving state is available. The each SMO reconstructs the expected fault in sensor based on discontinuous injection term used for converging output error to zero. Based on the reconstructed fault by each SMO, faults are detected using threshold approach. When the fault is detected, the reconstructed fault is used for fault tolerant control by subtracting to faulty data. The proposed fault detection (FD) and fault tolerant control (FTC) algorithms were evaluated using actual driving data and a three-dimensional (3D) vehicle model with a linear quadratic regulator for following control. The evaluation results are presented and analyzed with regard to fault reconstruction, detection, and tolerant control in four cases wherein two types of faults were applied.

show abstract

Section: Introductionmentioning

confidence: 99%

Dual-sliding mode approach for separated fault detection and tolerant control for functional safety of longitudinal autonomous driving

Song

Min

et al. 2020

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

show abstract

“…3 Lane detection fault from the vision sensors in commercial level, as covered in this study, may often occur because of internal/external factors. However, unlike with obstacle detection techniques using various sensors such as radar and lidar detectors, 4 it is difficult to design hardware redundancy for lane detection technology using various sensors because the alternative sensors used in lane detection technology are limited. For this reason, analytical sensor redundancy is more acceptable than hardware sensor redundancy when applied to lane detection function.…”

Section: Introductionmentioning

confidence: 99%

Robust lane tracking algorithm for forward target detection of automated driving vehicles

Song

Kim

et al. 2018

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

Detection of a primary forward target is one of the most important factors in a longitudinal control system of automated driving vehicles. If there is no special event such as a lane change, a lane detected by a vision sensor can be deemed as a predicted driving path of the ego vehicle. Based on this, it is possible to determine the state of the primary forward target vehicle accurately using the detected lanes. However, malfunction of vision sensors can induce unrecognized and misrecognized lane detection problems on account of internal/external environment factors. Furthermore, if a detection point is getting farther from the subject vehicle, the detection accuracy is getting lower because of inaccuracy of lane model parameters. To solve these problems, a novel method of lane tracking has been investigated. First, an integrated sensor module that combines a virtual sensor and a vision sensor has been developed. The virtual sensor is combined kinematic and dynamic vehicle model to be used in the full-speed range. Second, a lane estimator to improve lane detection accuracy at a long distance has been developed. The lane width in the public road can be assumed to be constant in the same road type. Based on this assumption, the clothoid parameters can be restored and consequently improve the lane detection accuracy. The performance of the proposed algorithm was verified by actual vehicle tests on public roads with manual driving and on proving ground with an automated driving system. The proposed algorithm has been compared with a conventional method which is based on in-vehicle yaw rate sensor. The test results have shown significant improvement of lane tracking performance over the conventional method.

show abstract

“…In another study, the Kalman filter and discrete wavelet transform were proposed as rain removal algorithms using the You Only Look Once (YOLOv3) methodology for the camera FDI of AVs [54]. Realpe et al [55] proposed a sensor fusion framework that integrates data from a unified configuration with sensor weight provided in a real-time FDI using the SVM algorithm to reduce the impact of sensor faults. Although Google, GM, BMW, and Tesla are trying and testing various AVs, multiple issues, such as bugs in traditional software, are a challenging task to fix, when compared with the DNN-based software [56].…”

Section: Phm Of Vision Sensorsmentioning

confidence: 99%

“…For example, even after applying Velodyne's correction factors and distance offset calibrated using readings from another reference LiDAR sensor, points with uncertainties of the order of 30 cm were noted [76]. A sensor fusion design that utilizes a support vector machine (SVM) technique to integrate data from a federalized fusion framework with sensor weight feedback data provided in real-time by the fault detection and diagnosis module was used to reduce the impact of sensor faults [55]. Duran et al [14] enlisted various kinds of faults in the LiDAR system, along with the severity level of fault occurrence in any particular component of the LiDAR system.…”

Section: Fault Categorymentioning

confidence: 99%

“…Noise in data due to weather condition [21] Unscented Kalman filter Displacements in sensor data [55] Support vector machine (SVM) algorithm Various faults (failure to read data, short-circuit, overvoltage, or misalignment) [14] Bayesian belief network approach Sensor data due to various environmental conditions (rain, fog, or lighting) [58] Deep neural network driven…”

Section: Fault Type Techniquementioning

confidence: 99%

See 1 more Smart Citation

Sensor-Based Prognostic Health Management of Advanced Driver Assistance System for Autonomous Vehicles: A Recent Survey

et al. 2022

View full text Add to dashboard Cite

Recently, the advanced driver assistance system (ADAS) of autonomous vehicles (AVs) has offered substantial benefits to drivers. Improvement of passenger safety is one of the key factors for evolving AVs. An automated system provided by the ADAS in autonomous vehicles is a salient feature for passenger safety in modern vehicles. With an increasing number of electronic control units and a combination of multiple sensors, there are now sufficient computing aptitudes in the car to support ADAS deployment. An ADAS is composed of various sensors: radio detection and ranging (RADAR), cameras, ultrasonic sensors, and LiDAR. However, continual use of multiple sensors and actuators of the ADAS can lead to failure of AV sensors. Thus, prognostic health management (PHM) of ADAS is important for smooth and continuous operation of AVs. The PHM of AVs has recently been introduced and is still progressing. There is a lack of surveys available related to sensor-based PHM of AVs in the literature. Therefore, the objective of the current study was to identify sensor-based PHM, emphasizing different fault identification and isolation (FDI) techniques with challenges and gaps existing in this field.

show abstract

Multi-sensor Fusion Module in a Fault Tolerant Perception System for Autonomous Vehicles

Cited by 8 publications

References 9 publications

Dual-sliding mode approach for separated fault detection and tolerant control for functional safety of longitudinal autonomous driving

Dual-sliding mode approach for separated fault detection and tolerant control for functional safety of longitudinal autonomous driving

Robust lane tracking algorithm for forward target detection of automated driving vehicles

Sensor-Based Prognostic Health Management of Advanced Driver Assistance System for Autonomous Vehicles: A Recent Survey

Contact Info

Product

Resources

About