Health Monitoring System for Autonomous Vehicles using Dynamic Bayesian Networks for Diagnosis and Prognosis
- PDF / 1,222,454 Bytes
- 21 Pages / 595.224 x 790.955 pts Page_size
- 84 Downloads / 192 Views
Health Monitoring System for Autonomous Vehicles using Dynamic Bayesian Networks for Diagnosis and Prognosis ˆ Gomes1 Iago Pacheco
· Denis Fernando Wolf1
Received: 30 June 2020 / Accepted: 26 November 2020 © Springer Nature B.V. 2020
Abstract Autonomous Vehicles have the potential to change the urban transport scenario. However, to be able to safely navigate autonomously they need to deal with faults that its components are subject to. Therefore, Health Monitoring System is a essential component of the autonomous system, since allows Fault Detection and Diagnosis. In addition, Prognosis System is also important, since it allows predictive maintenance and safer decisions during vehicle navigation. This paper presents a Hierarchical Component-based Health Monitoring System with Fault Detection, Diagnosis and Prognosis using Dynamic Bayesian Network (DBN) with residue generation, a combination of knowledge-based and model-based detection, diagnosis and prognosis approaches. We evaluate the proposed Dynamic Bayesian Network using different machine learning metrics and a dataset with sensor readings gathered using the CaRINA II autonomous vehicle platform, and the CARLA simulator. Both simulated and experimental results demonstrated a positive performance of the DBNs even with high rate of missing data for some of the model’s variables. Keywords Health monitoring system · Prognosis system · Fault detection and diagnosis · Autonomous vehicle · Dynamic bayesian network
1 Introduction Autonomous Vehicles have the potential to change the urban transport scenario improving safety, mobility, reducing traffic congestion, pollutant emission, fuel consumption, land usage, among other positive societal impacts [3, 4]. To do so, they consist of many sub-systems responsible for sensing, planning, and controlling. First, raw sensor data This study was financed in part by the Coordenac¸ a˜ o de Aperfeic¸oamento de Pessoal de N´ıvel Superior - Brasil (CAPES) Finance Code 001 and grant 88887.500344/2020-00, the Brazilian National Research Council (CNPq) under grant 166874/2017-5, and the S˜ao Paulo Research Foundation (FAPESP) under grant 2019/27301-7. This paper was presented in part at PROCEEDINGS OF THE IEEE 2019 19th International Conference on Advanced Robotics (ICAR) [1] and IEEE 2018 Latin American Robotic Symposium (LARS) [2]. Iago Pachˆeco Gomes
[email protected] Denis Fernando Wolf [email protected] 1
Institute of Mathematics and Computer Science Laboratory of Mobile Robotics, University of S˜ao Paulo, S˜ao Carlos, Brazil
provides what it takes for understanding the surrounding environment through perception algorithms, which supply decision-making and planning components, to finally generate control outputs safely moving the vehicle towards its destination [5]. The autonomy of autonomous systems is directly related to the number of tasks they are able to perform without human intervention. The International Standard SAE-J3016 classifies the autonomy levels in 6 categories, from 0 (no-automation) to 5 (fully-autonomou
Data Loading...
