2020 IEEE Transactions on Intelligent Vehicles  
In the context of autonomous driving, where humans may need to take over in the event where the computer may issue a takeover request, a key step towards driving safety is the monitoring of the hands to ensure the driver is ready for such a request. This work, focuses on the first step of this process, which is to locate the hands. Such a system must work in real-time and under varying harsh lighting conditions. This article introduces a fast ConvNet approach, based on the work of original work
more » ... of OpenPose by Cao, et al. for full body joint estimation. The network is modified with fewer parameters and retrained using our own day-time naturalistic autonomous driving dataset to estimate joint and affinity heatmaps for driver and passenger's wrist and elbows, for a total of 8 joint classes and part affinity fields between each wrist-elbow pair. The approach runs real-time on real-world data at 40 fps on multiple drivers and passengers. The system is extensively evaluated both quantitatively and qualitatively, showing at least 95% detection performance on joint localization and arm-angle estimation. Control of whole-body vibration (WBV) via a seat suspension in off-road vehicles is a challenging task due to the presence of severe external disturbances and parametric uncertainties. In this paper, a novel optimal robust (mixed H ∞ /H 2 ) controller is proposed to achieve enhanced vibration attenuation performance of seat suspensions considering the parametric uncertainties due to variations in driver mass and external disturbances encountered at the cabin floor including occasional shocks. A direct Lyapunov based LMI approach is employed to prove stability of the closed-loop system. In search of an optimal solution for the designed robust controller, a H ∞ disturbance attenuation performance together with the weighted H 2 norm for minimizing the mean disturbance rejection, are considered using the Chaos-enhanced Accelerated Particle Swarm Optimization (CAPSO). The optimization problem for the proposed controller is formulated to minimize the frequency-weighted vibration dose value (VDV) due to acceleration response at the seat, while constraining the relative displacement between the seat and the seat base. The effectiveness of the proposed controller is illustrated through comparisons with performance achieved through some of the reported other seat suspension robust controllers in addition to a passive suspension. Results show that the proposed optimal robust controller could provide substantial reductions in the frequency-weighted acceleration at the seat and the VDV, measures of the force applied to the driver due to terrain induced vibration and shock, while limiting the relative displacement between the seat and the seat base.
doi:10.1109/tiv.2020.2978681 fatcat:n7ifvfboe5crbdlqykb7rfwgk4