Human-vehicle interaction in automated vehicles: the issue of carsickness

Motion Sickness (MS) is an issue of most transportation systems. Several countermeasures to this problem in cars are proposed in the literature, but most of them are qualitative, behavioural or involving complex chassis systems. Autonomous Driving (AD) can exacerbate the problem of MS due to the change from driver to passenger with the consequent loss of control over the vehicle. With the growing interest in self-driven vehicles, the issue of MS may be so important as to undermine their benefits in terms of increased productivity; not addressing this issue may limit the users’ acceptance reducing the safety and the environmental impact of autonomous vehicles. In this thesis, the issue of carsickness is discussed, analysing the potential technologies to monitor MS in cars and discussing their feasibility. After the analysis of the monitoring technologies, in the final part of the manuscript, the issue of reducing MS in cars is analysed, proposing optimal methods for carsickness reduction. To optimise the vehicle behaviour, two Model Predictive Control (MPC) problems are formulated to analyse the potential impact in different applications of optimal MS reduction techniques. The first task is to find the optimal speed profile to travel on a given path while trading-off between minimal travel time and minimal Motion Sickness Incidence (MSI); in this part, several strategies are proposed and analysed comparing their performance. This novel methodology may be used in autonomous cars to create a reference velocity profile for lower control layers or, in human-driven ones, to advise the driver. The second task applies only to autonomous vehicles, implementing the MS optimisation in the motion planning layer; a trajectory optimisation problem is solved using the best performing strategies of the optimal speed profile task. The results show that optimising vehicle behaviour may significantly reduce the MSI, improving the user experience in cars; furthermore, the wider possibilities offered by autonomous cars allow for further reduction of MSI. The optimal methodologies proposed and the strategies used are the main contribution of this doctoral thesis; the coherency of the results in the different cases analysed suggests that these strategies have a general validity for MS reduction. In the conclusions chapter, the impact of these methods is discussed; possible ways of integrating MS monitoring technology into the proposed reduction techniques are analysed.