Distributed and Adaptive Shared Control Systems: Methodology for the Replication of Experiments

2015 ◽  
Vol 22 (4) ◽  
pp. 137-146 ◽  
Martin F. Stoelen ◽  
Virginia Fernandez de Tejada ◽  
Alberto Jardon Huete ◽  
Carlos Balaguer ◽  
Fabio Paolo Bonsignorio
2018 ◽  
Vol 48 (5) ◽  
pp. 509-525 ◽  
David A. Abbink ◽  
Tom Carlson ◽  
Mark Mulder ◽  
Joost C. F. de Winter ◽  
Farzad Aminravan ◽  

Seshan Ramanathan Venkita ◽  
Dehlia Willemsen ◽  
Mohsen Alirezaei ◽  
Henk Nijmeijer

One of the main safety concerns associated with semi-autonomous vehicles is the sharing of control between a human driver and an autonomous driving system. Even with an attentive driver, such switches in control may pose a threat to the safety of the driver and the surrounding vehicles. The aim of this study is to develop an indicator that can measure the level of safety during a driver take-over, using knowledge about the system known a priori. A model-based approach is used to analyse the system with special focus on the lateral dynamics of the vehicle. The driver and the vehicle are modelled as linear systems, and a path tracking controller is used to serve as an autonomous system. With this structure, shared control is studied as a switched system, in which the vehicle’s lateral control switches between the autonomous system and the driver. A bound on the transient dynamics that arise due to a switch is derived, using the induced [Formula: see text] norm. This bound is then used to formulate an indicator that checks if the states/outputs of interest are within acceptable limits. A comparison with simulation results has shown that the indicator successfully captures the effect of different system parameters on take-over safety, although in a slightly conservative manner. This indicator can be further developed as a tool to be used in the design and evaluation of shared-/multi-modal control systems in future vehicles.

2018 ◽  
Vol 66 (2) ◽  
pp. 146-159 ◽  
Julian Ludwig ◽  
Manuel Martin ◽  
Matthias Horne ◽  
Michael Flad ◽  
Michael Voit ◽  

Abstract In the near future, drivers of automated cars will still have to take over from time to time at short notice. Current control systems implement a hard switch, disabling the automation all at once. However, studies show that the driver’s ability to take over depends on his last activity. We therefore propose a system that uses camera based observation of the driver to assess the situation and to predict transition times. We combine this with a control system that uses a cooperative shared control method to support the driver in takeover situations and allows him to adjust safely to the current situation. We present our first steps towards this goal and show both how the behavior of the driver in the interior can be assessed and how a cooperative control transfer can be implemented. We further point out the necessary steps to implement the proposed system and give a first impression of the performance via simulation.

Sign in / Sign up

Export Citation Format

Share Document