In order to guide the vehicle autonomously, an appropriate navigation concept is required that supplies the vehicle's pose in terms of position, velocity and attitude. Thereby, the navigation solution has to fulfil requirements on accuracy, bandwidth, availability, integrity and smoothness. In this regard, this contribution assumes that the requirements regarding availability and integrity are always met. For the feasibility of autonomous evasion maneuvers, the upper bound of the admissible horizontal position error is assessed to be 0.5 m. As the low-cost navigation solution does not have the ability to fulfil this requirement when considering absolute positioning, relative positioning with respect to the initial maneuver position is investigated. Particularly, the navigation concept relies on a loosely coupled GNSS/INS integration concept that complements the properties of an INS (high update rate, sensor drift) and a GNSS (low update rate, no sensor drift). Moreover, a real-time capable solution to cope with time delayed GNSS measurements is introduced and an extension to aid and thus to improve the yaw angle estimate is proposed. Though, larger position errors and even disadvantageous discontinuities (stepwise changes in the horizontal position) can be observed when operating the vehicle at the handling limits. These errors are either caused by discontinuities in GNSS measurements or result from simplifications in the filter design (e.g. for real-time compensation of time delays). To solve this issue, it is proposed to decrease the weight on GNSS position and velocity measurements appropriately in the filter update step during the maneuver. Experimental results show that all the requirements on the navigation solution are satisfied and a convincing estimation performance is achieved. Particularly, a maximum horizontal position error of 0.24 m is obtained for a maneuver at the handling limits which is significantly less than the specified feasibility limit. Thus, the navigation concept is assessed to be appropriate for autonomous evasion maneuvers at the vehicle handling limits.


    Access

    Access via TIB

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Optimal vehicle dynamics control and state estimation for a low-cost GNSS-based collision avoidance system


    Contributors:


    Publication date :

    2014


    Size :

    208 Seiten, 65 Bilder, 14 Tabellen, 98 Quellen




    Type of media :

    Theses


    Type of material :

    Print


    Language :

    English