A lane departure detection method is proposed using Takagi-Sugeno (T-S) fuzzy approach in this paper. To deal with this problem, a nonlinear model deduced from a vehicle lateral dynamic and a vision system is represented by an uncertain T-S fuzzy model affected by unknown inputs. The developed lane departure detection technique is based on an unknown inputs fuzzy observer. Design conditions of such observers are expressed in terms of Linear Matrix Inequalities (LMI). Indeed, the road curvature considered as unknown input is estimated and compared to the vehicle trajectory curvature. The proposed algorithm allows to reduce false alarms and to integrate the driver corrections by taking the steering dynamics into account. In order to show the efficiency of the given method, simulations with different driving scenarios are proposed. Based on T-S fuzzy modelling and road curvature estimation, a technique of lane departure detection is developed. The method assumes that only one sensor is used and no knowledge of the path road is needed. The nonlinear model obtained from a vehicle lateral dynamics and a vision system is represented by an uncertain T-S fuzzy system. The road curvature is considered as an unknown input. Then an unknown inputs T-S observer has been designed to estimate system both vehicle states and road curvature. The design conditions are given in LMI terms easy to solve using existing numerical tools. The algorithm proposed to detect lane departures is very efficient and practical; it uses two risk indicators and takes into account the steering dynamics. Indeed, the difference between the two curvatures is used as the first driving risk indicator. To reduce false alarms, the second driving risk indicator integrates the driver corrections and introduces the time to the lane keeping to anticipate lane departure detections. The efficiency of the proposed risk indicators is shown by considering two driving scenari in a double lane change.


    Access

    Access via TIB

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Design of unknown inputs robust fuzzy observer for lane departure detection


    Contributors:
    Dahmani, H. (author) / Chadli, M. (author) / Rabhi, A. (author) / Hajjaji, A. el (author)

    Published in:

    Publication date :

    2011


    Size :

    17 Seiten, 11 Bilder, 1 Tabelle, 19 Quellen



    Type of media :

    Article (Journal)


    Type of material :

    Print


    Language :

    English