… Using an exploratory approach with a qualitative research design, we conducted semi-structured
interviews to reveal relevant autonomous driving acceptance criteria from an end-user …

… Although there are some technical differences between RL and Deep RL, for the purposes
of this review, which is to differentiate IL and RL, we will use the terms RL and Deep RL …

… It assigns a weight (positive or negative score) to each trajectory based on parameters defined
by the end user, such as safety, distance traveled, comfort, and others. Rhinehart et al. [77…

… We define end-to-end autonomous driving systems as fully differentiable programs … use
RL in autonomous driving have only investigated the technique in simulation. Most works use …

… [13] were the first to propose a successful use of imitation learning for autonomous driving.
Bojarski et al [6] predict steering commands for lane following and obstacle avoiding tasks. …

… CONCLUSION In this paper, we explored the use of uncertainty in endto-end controllers for
self-driving cars and suggested new evaluations for it. We studied two separate architectures…

… Where this differs from traditional user requirement gathering is that the end user is a
machine. Maps for machines differ from maps designed for humans, as they can contain much …

… It is also evident that autonomous driving as a safety-critical … open-loop end-to-end
autonomous driving methods. We … that none of the existing methods use the historical trajectory …

… app enables both companies to study the end-to-end user experience of on-demand hailing
… user experience research around how passengers book and interact with autonomous cars. …

… Moreover, autonomous driving systems should be able to predict … an end-to-end driving
system especially when the speed … We use Carla’s autopilot to collect the data for observations, …