Embedded Bayesian Network Contribution for a Safe Mission Planning of Autonomous Vehicles

Catherine Dezan,Sara Zermani,Chabha Hireche

doi:10.3390/a13070155

Abstract

Bayesian Networks (BN) are probabilistic models that are commonly used for the diagnosis in numerous domains (medicine, finance, transport, robotics, …). In the case of autonomous vehicles, they can contribute to elaborate intelligent monitors that can take the environmental context into account. We show in this paper some main abilities of BN that can help in the elaboration of fault detection isolation and recovery (FDIR) modules. One of the main difficulty with the BN model is generally to elaborate these ones according to the case of study. Then, we propose some automatic generation techniques from failure mode and effects analysis (FMEA)-like tables using the pattern design approach. Once defined, these modules have to operate online for autonomous vehicles. In a second part, we propose a design methodology to embed the real-time and non-intrusive implementations of the BN modules using FPGA-SoC support. We show that the FPGA implementation can offer an interesting speed-up with very limited energy cost. Lastly, we show how these BN modules can be incorporated into the decision-making model for the mission planning of unmanned aerial vehicles (UAVs). We illustrate the integration by means of two models: the Decision Network model that is a straightforward extension of the BN model, and the BFM model that is an extension of the Markov Decision Process (MDP) decision-making model incorporating a BN. We illustrate the different proposals with realistic examples and show that the hybrid implementation on FPGA-SoC can offer some benefits.

Highlights

Bayesian Networks (BNs) are well known probabilistic models to elaborate diagnosis taking into account uncertainties
We propose a complete design flow from failure mode and effects analysis (FMEA)-tables towards FPGA implementation dedicated to the BN monitors
For Bayesian networks based on FMEA, we propose an architecture where the parameters of the BN are stored in BRAM and can be modified using an AXI_BRAM Controller, which manages the data transfer between the BRAM blocks and the DDR memory

Summary

Introduction

Bayesian Networks (BNs) are well known probabilistic models to elaborate diagnosis taking into account uncertainties. Numerous fields of applications like medicine, robotics, industrial process, exploit their inference mechanism to evaluate the causes of a disease or the source of the malfunctioning from the observed environment. With the autonomy of unmanned aerial vehicles (UAV), the safety becomes a major issue to achieve the mission correctly. The analysis and the evaluation of the dependability of the system [1] including the safety and the security constraints are crucial. In the case of aerospace systems, a set of faults and errors [2] related to sensors, actuators and embedded systems can be identified in advance. To detect and to mitigate the different kinds of hazards, number of model-based fault detection isolation recovery (FDIR) techniques have been introduced [3]

Methods

Discussion

Conclusion