ix
List of Figures
1.1 Autonomous vehicle functional units. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 e six levels of autonomy in road vehicles, as defined by the Society of
Automotive Engineers (SAE). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.1 Deep learning is a subset of machine learning techniques. . . . . . . . . . . . . . . . . . 5
2.2 Feedforward neural network. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 Common activation functions in neural networks. . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 Recurrent neural network, where the red arrows represent temporal
connections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.5 Convolutional neural network.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.6 Reinforcement learning process. Agent observes states s
t
, takes action a
t
,
receivers reward r
t
, and new observation s
t C1
. e transition
{s
t
; a
t
; r
t
; s
t C1
} is then used to update the network parameters. . . . . . . . . . . . 11
2.7 An actor-critic agent interacting with the environment. Dashed lines
represent a parameter update.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1 An example of a goal structuring notation for safety case argumentation of
a deep neural network, where the boxes illustrate the goals, rounded boxes
provide context for a goal, and rhomboids denote strategies to fulfill the
goals. Adapted from [179]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.