HCI
CARS

Automated
Vehicles


Interaction
Methods

Speech

Touch
Screen

Gestures

Context of Use

🔘 Physical
Controls

Advantages

Eyes on road, intuitive, delight factor

Disadvantages

learn the gestures, hands of wheel, limited set

Advantages

Natural, keep hands an eyes on driving

Disadvantages

Social context, error recovery, phrases that
that have to be learnt

Advantages

Intuitive, increased functionality, flexible

Disadvantages

Lack of tactile feedback, visual search, menu navigation, screen smudges

Advantages

Disadvantages

Evaluation
Methods

In-Vehicle
Technology

Non-User

User

Social
Context

Mental
Workload

How these secondary tasks impact
on the safety of the primary task of driving

Familiar, tactile feedback

Clutter potential,
restrictions on design,
costly to make changes

Simulator

Occlusion

Cooperative Usability evaluation
use of scenario & verbal protocol

Heuristics

Expert Assessment

Modelling

Eye Tracking

Advantages

  • Occlusion

  1. Occlusion is a user trial method which aims to predict visual demand of user interfaces. i.e. Determine how well users are able to complete a task in the absence of vision.


    Inference


    More vision time = higher workload/


    Less vision = less workload.


  2. Participants complete tasks in a simulator whilst wearing LCD goggles that provide a cycle of vision/non vision (Cycle of 1.5 secs open/1.5 secs closed).


  3. The method assesses how easy it is to achieve tasks with an interface under conditions of interrupted vision (as would be the case in a moving vehicle).



Key metrics are:
Total Shutter Open Time (TSOT=amount of vision needed); Resumability Ratio (TSOT/time needed with full vision).

KLM

  • Combination of occlusion with KLM

    KLM: it is an analytical method for predicting task times with an interface; tasks are broken down into fundamental operators (e.g. keying, pointing, mentally preparing);
  • Mean time values are available for different operators;
  • Predicted task time is calculated by describing operators and summing together average times;
  • The method assumes errorFfree performance.
    KLM can be used to predict the metrics of occlusion using basic assumptions that consider what aspects of the task can/cannot be done with vision.
    For example, when considering an in vehicle infotainment task (e.g. changing a music track) mental and pointing operators associated with locating a finger in a touchscreen will need vision, whereas pressing a button repeatedly once the finger is located may not.

Space

Where will all new features go in such a limited space

Large-user
group

People are less likely to engage with a speech device

Will differ on cognitive ability

Environment

Highly variable - temp, light, sound

Underload


  • Levels of Automation

  • 0 None
  • 1 Function Specific
  • 2 Combined Function
  • 3 Limited self-driving
  • 4 Full self driving

  • Disadvantages:

    • load of driving task
    • situation awareness
    • fundamental driving skills
    • negative behavioural adaptation
  • Lack of system acceptance
  • monitoring errors e.g not realising the system is faulty or off

  • Advantages

  • Safety
    • A lot of human error taken out & will reduce the number of accidents
  • Efficiency
    • Optimal traffic flow & reduced emissions
  • Comfort
    • can stretch out on long highway journeys
  • Productivity
    • Driver is handsfree to focus on other activities

Safe & controlled environment - to do loading tasks
Cost effective way to get data

Disadvantages

Question of validity - do people drive the same in a simulator
Element of Invincibility - never going to really crash
Perception of speed will be different

Mirrorless Cars

Advantages

Disadvantages

  • HCI Issues

    How to design for automated driving functionality
    • management of transition periods and or warnings
    • Shared situation awareness for vehicle / driver
      How to design for non-automated driving functionality
    • Freedom for different vehicle interiors
      How to design for whole Human Machine Interface
    • how will it communicate with other road users

WB!

  • Problems in road transport have led to moves to develop more intelligent cars
    Potential for these systems to improve traffic safety, efficiency and driver comfort
  • But because of the complexity in the driving task there are many concerns
  • Therefore HF/HCI issues have to be central to the design process for this new technology

Overload

  • High levels of mental workload
  • Attention is going to be divided,
  • Safety put at risk
  • Increased stress