Lecture 6 - Automated Vehicles Flashcards
Level 0 car
There are no autonomous features
Level 1 car
These cars can handle one task at a time, automatic braking
Level 2 car
These cars would have at least two automated functions
Level 3
These cars handle “dynamic driving tasks” but might still need intervention
Level 4 car
These cars are officially driverless in certain environments
Level 5 car
These cars can operate entirely on their own without any driver present
Afraid of riding a fully autonomous vehicle statistics
2017: 63%
2018: 73%
2019: 71%
2020: 86% (of which 32% “unsure”)
Outsized reactions
Outsized reactions in the face of inevitable accidents
“They said they need real-world examples, but I don’t want to be their real-world mistake”
Human Machine Interface
- Vehicle and human can be seen as a
joint cognitive system - The HMI provides continuous
interaction between human and vehicle
3 main challenges HMI
- Minimize mode errors
- Stimulate appropriate level of attention
and intervention - Minimize automation surprises
Minimise mode errors
Can the automated system…
* Handle longitudinal and lateral control
* Manoeuvre in the current environment and situation (e.g. handle a roundabout)?
* Perform strategic aspects of driving (e.g switch lanes or change route)?
Stimulate appropriate levels of attention and intervention
Level 2: Driver attention to the road needs to be as high as during manual driving
The higher the reliability of the system, the lower will be the attention of the driver and the higher the impact of a failure
Minimize automation surprises
Two types:
* Absence of expected action
* Presence of unexpected action
eHMI (external-Human Machine Interface) - main challenges
- Communication
- Scalability
- Colour
eHMI Communication - instruction or intention?
Intention
The colour of eHMI
Cyan!
- Compliance with road regulations
- Red, green, yellow, and blue can generate confusion
Ethic dilemma’s
How should an autonomous vehicle “ethically” behave?
Trust in automation
- Uncertainty and vulnerability
- Not unidimensional construct
- Situation specific
Trust
Trust is “the attitude that an agent will help achieve an individual’s goals in a situation characterized by uncertainty and vulnerability”
Dispositional trust
Overall tendency to trust automation, depends on:
- Culture
- Age
- Gender
- Personality
Situational trust
Trust experienced in a specific situation
Depends on:
1. External variability
* Road conditions
* Weather conditions
2. Internal variability
* Self-confidence
* Mood
(Dynamic) learned trust
Trust that users develop over time, based on the skills and knowledge acquired through past experiences and interactions with the system
High trust ≠ Appropriate trust
Design for appropriate trust, not greater trust!
Trust calibration - Guidelines
- Design for appropriate trust, not greater trust
- Show the purpose of the automation
- Train operators regarding its expected reliability
- Carefully evaluate any anthropomorphizing of the automation
