GATEway

From FOT-Net WIKI
Jump to: navigation, search
Contact details

Simon Tong TRL stong@trl.co.uk

Timing and duration of tests

3 trials to be conducted within 2016

Location(s) of tests

Greenwich, London, UK

Tested automation

Tested functionalities

Fully automated passenger shuttles; valet parking functions; automated local delivery services

Level of automation tested

SAE: Trial 1, passenger shuttles, levels 4-5; Trial 2, valet parking, levels 3-4; Trial 3, automated delivery service, levels 3-5. Levels of automation still to be confirmed as vehicle procurement and specification is underway.

Tested use cases

Trial 1: ‘last mile’ public transportation shuttle between key transport hubs and local residences/businesses. Serving a variety of passenger sub-groups including people with disabilities. Trial 2: demonstration of an automated valet parking service to explore public perception and likely adoption in different forms/scenarios. Trial 3: demonstration of an automated local delivery service to explore suitability of systems to cover local courier-style deliveries for nearby businesses.

Tested transport system

Trial 1: 6-person passenger shuttle, designed specifically for last mile transportation in urban areas, ideally using pedestrian/cycle routes Trial 2: modified M1 passenger vehicle with automated parking function Trial 3: N1 vehicle for carriage of goods designed specifically for automated operation

Purposes of testing
  • Assessment of driver behaviour (human, vehicle) / road user behaviour
  • Assessment of impacts on safety (short-term, long-term)
  • Assessment of user (driver, traveller, etc.) acceptance, usability, take up, etc.
  • Assessment on transport system, infrastructure, built environment, land use, other societal factors (e.g. societal integration), etc. (long-term)
  • Interaction with other traffic participants (automated vehicles, conventional vehicles, vulnerable road users)
  • Technical assessment, proof of concept (incl. vehicle, background support systems such as communication)
  • develop technology-agnostic test environment, procedures and standards

Methodology

Definition of baseline

No formal ‘baseline’ condition as such; however, comparisons will be drawn with existing transport provisions when inviting feedback on automated systems. Some specific activities will have baseline conditions – e.g. pedestrian interactions with automated shuttles will be contrasted with interactions that occur with conventional vehicles.

Test design

Majority of testing occurs across Trials 1-3.

Design for Trial 1: passenger shuttle ‘service’ running in up to two locations in Greenwich, London. Providing links between transport hubs and businesses/residences/tourist destinations. Trial will have intensive data collection period of approximately 3-4 weeks. During this period, data will be collected using surveys, focus groups, observations, interviews, telematics and social media networks. Different shuttle services will be tested, including on-demand routing. Specific user groups will be engaged, including people with disabilities. Data will explore interactions between shuttles and pedestrians/cyclists, reactions to the shuttle, experiences of passengers and members of the public, future scope for implementation and expansion.

Design for Trial 2: currently proposed as a valet parking demonstration in which members of the public will experience the vehicle self-parking from a valet drop-off point. Data will be collected using survey and focus group methods. Different use cases will be explored.

Design for Trial 3: currently proposed as an automated delivery service. Data likely to be collected using similar methods to Trial 1. Samples surveyed will include local businesses and customers that may interact with the delivery shuttle. Expected to test boundaries of level 5 automation in a local controlled environment.

Additional simulator trials based on simulated models of the Greenwich environment.

Method of testing

Controlled field tests, Driving simulator, Focus groups, Interviews, Naturalistic field tests, Questionnaires, Simulation, Test track


Test fleet, participants and environment

Number and make of vehicles

Trial 1: 7x passenger shuttles (Westfield/Heathrow ‘pods’) Trial 2: 2x M1 valet parking passenger vehicles Trial 3: 1x N1 delivery vehicle

Description and number of participants/drivers

Trials 1-3: number of participants TBC

Tested environment and facilities

Greenwich Peninsula – pedestrian routes for passengers shuttles. Mixed public/private road network and car parks for further testing of automated functions.

Wider Greenwich locality – mix of tourist/heritage sites with suitability for automated vehicle trials. Local need to deliver improved transport links to major public transport hubs located to the north of the borough.

Legal and ethical aspects

Comprehensive safety case and standards being developed to accompany trials. Safety standards expected to form publishable deliverables to supplement UK DfT code of practice for automated vehicle testing.

Duration of testing

Each trial has a concentrated test period of 3-4 weeks. Trial 1 expected to have a longer 6 month demonstration period following intensive testing. Trial 1 shuttle routes are expected to cover approximately 3km. Other test routes TBC.

Input parameters and assumptions of simulation tests

TBC.

Data

Logging
  • CAN
  • Communication data (V2V, V2I, V2U, etc.)
  • GPS
  • HMI
  • In-vehicle cameras
  • Map data
  • Other cameras
  • Sensor data
  • Situational data (e.g. weather, traffic situation)
  • System internal data
Specifications for the data sources
  • Other cameras, specify: external facing cameras and cameras monitoring the test environment
  • Communication data (V2V, V2I, V2U, etc.), specify: TBC, but V2V and V2I expected as minimum
  • System internal data, specify: TBC based on vehicle design and
  • Sensor data, specify: TBC based on vehicle design and capability
  • Situational data (e.g. weather, traffic situation), specify: weather, pedestrian activity
Key Performance Indicators (KPIs)

Trial 1: Frequency and type of shuttle interactions; passenger feedback; frequency and type of interventions by the shuttle guardian; levels of route complexity traversed successfully at the highest level of automation; highest journey speed delivered safely. Trial 2: TBC Trial 3: TBC

Situational data available

Trial 1: weather, route characteristics, volume of pedestrian and/or cyclist activity, time of day Trial 2: weather, parking characteristics, volume of pedestrian and/or cyclist activity, time of day Trial 3: TBC

Subjective data collected

Trials 1-3: participant experiences of and interactions with automated vehicles; bystander experiences of and interactions with automated vehicles; stated preferences; relevance measures for future transport systems

Results

Issues that affected the impact assessment

N/A

Results

N/A

Publications

N/A

Other things to report

Facts about "GATEway"
Method of testingControlled field tests +, Driving simulator +, Focus groups +, Interviews +, Naturalistic field tests +, Questionnaires +, Simulation + and Test track +
Purpose of testingAssessment of driver behaviour (human, vehicle) / road user behaviour +, Assessment of impacts on safety (short-term, long-term) +, Assessment of user (driver, traveller, etc.) acceptance, usability, take up, etc. +, Assessment on transport system, infrastructure, built environment, land use, other societal factors (e.g. societal integration), etc. (long-term) +, Interaction with other traffic participants (automated vehicles, conventional vehicles, vulnerable road users) + and Technical assessment, proof of concept (incl. vehicle, background support systems such as communication) +