ADVISORS

1
WWW.ADVISORS.IAO.FHG.DE
2
What can ADVISORS do for freight transport?
WWW.ADVISORS.IAO.FHG.DE
3
This presentation

• ADVISORS Objectives and partners
• Actors interests and acceptance
• Risk analysis
• Evaluation methodology
• Conclusions

4
ADVISORS

• Advanced Driver assistance and Vehicle control
  systems Implementation, Standardisation, Optimum
  use of the Road network and Safety
• DGTREN GRD1-2000-10047
• April 2000 October 2002

5
Main Objectives

• Integrated assessment methodology criteria
• Identify ADAS with high potential for safety and
  network efficiency increase and environmental
  load reduction
• Comprehensive Implementation Scenarios

6
Partners

• SWOV
• JAM DE RIJK BV
• ACHMEA Holding BV
• Delft University of Technology
• IBSR/BIVV
• Aristotle University of Thessaloniki
• VTI
• University of Groningen

• CRF
• Siemens Automotive SA
• University of Stuttgart
• Bundesanstalt fur Strassenwesen
• National Technical University of Athens
• Centrum Dopravniho Vyzkumu
• VTT
• TRL

7
Actor interests and acceptance
8
Actor stated problems
2000
2010
6.8
Road network and its costs
7.7
6.5
Fleet and resources and their costs
6.8
6.6
Accessibility of public transport
6.8
Travel time, its predictability and information
6.8
6.6
7.1
Traffic safety
6.4
6.8
Environmental effects of traffic
6.7
1
2
3
4
5
6
7
8
9
10
9
Problems, different actor groups
10
9
8
7
6
5
4
Environment
Traffic safety
Road network
Travel times
Accessibility of
Fleet and
and its costs
public transport
resources
Key actors
Authorities
Industry
Fleet managers
Other important actors
10
Difference between mean severity of
traffic-related problems currently and in 2010.
11
Attractiveness on motorways
12
Results

• Most severe rated safety problems
• Increasing network related problems and costs
• Dynamic navigation function most attractive
  (comfort, travel time, fuel consumption)
• Trucks ACC (safety, comfort)
• ISA mostly urban (safety, comfort for trucks)
• Warning more attractive than intervention
• User friendly terminology

13
Risk analysis
14
Based on FMEA (Failure Mode and Effects Analysis)
methodology

• The indicators of severity, occurrence
  probability, delectability and recoverability
  have been expanded to cover technical-,
  behavioural-, legal- and organisational risks.
• For each identified type of problem
• failure effect,
• cause,
• detection and recognition,
• mitigation strategy

15
Some elements of the Methodology for ADAS
multiparameter risk analysis

• Risk Severity x Probability x ((Detectability
  Recoverability)/2)
• Calculate Technical Risk Number (RNT),
  Behavioural Risk Number (RNB), Legal Risk Number
  (RNL) and Organisational Risk Number (RNO).

16
Technical Failure Analysis steps

• System definition
• Development of block diagrams
• Failure modes, causes and effects
• Design alternatives
• Criticality of failure mode
• Probability of failure mode
• Criticality evaluation

17
Severity levels of technical failure

• Extremely severe Technical failure is
  intolerable it could cause loss of vehicle
  safety resulting in serious injury and vehicle
  or environmental damage.
• Severe The failure implies total loss of the
  vehicle availability causing major customer
  dissatisfaction.
• Moderate Failure implies the partial loss of the
  function causing customer dissatisfaction.
• Slight Technical failure of ADAS could cause
  customer dissatisfaction without resulting in
  injury or vehicle or environmental damage.
• Insignificant The technical failure is very
  unlikely to happen and the consequences would be
  insignificant.

18
Technical barriers to ADAS deployment

• A technical solution is not available, needs
  further investigation, or is highly complicated.
• Cost would be prohibitive.
• The benefits gained from the functionality of the
  system are uncertain.

19
Technical Risk Analysis results

• Extended Navigation System (ENS)
• Route guidance, navigation and vehicle status
  system (mainly for truck applications) (RGNVSS)

20
Extended Navigation (EN)

• Navigation map enriched with additional
  information, speed limits in each road segment,
  road curvature in each road segment, number of
  lanes in each road segment, landmarks (traffic
  signs) in each road segment. This additional
  information can be provided to the driver
  directly, without filtering, or can be used in
  ADAS application.

21
Conclusions EN

• The info provided may be wrong or outdated (road
  lanes, directions and other road characteristics
  do change overnight due to local traffic
  policies, road works, etc). System reliability
  depends heavily upon data mapping from on-board
  sensors as well as frequent or even dynamic map
  updates.

22
Preliminary guidelines for EN

• Always fusion between extended map data and
  on-board sensors (from radar, road lane
  recognition camera, etc.) before providing
  information or warning to the driver.
• In case of conflict, the on-board sensor data
  should receive priority.
• A driver feedback button should be provided.
• The development and use of dynamically (locally)
  updated databases in the future is recommended.

23
To be evaluated (EN)

• Define the optimal strategy and HMI of driver
  feedback, simulating various system errors /
  inaccuracies,
• Define the optimal combination of redundant
  driver warning modalities.

24
Route guidance, navigation and vehicle status
system (mainly for truck applications) (RGNVSS)

• Collection and recording of data related to
  trip, driver, vehicles location, performance,
  automatically or by driver (i.e. reasons for
  delay, expenses, etc).
• Messages receipt and transmission between vehicle
  and base, for route guidance and rerouting of the
  vehicle. Based upon INMARSAT-C and Euteltracs
  satellite communications and GSM-sms messaging.

25
RGNVSS contd

• Exchange of transport orders and contract status
  between the vehicle, its base and clients
  (vehicle rerouting / operations rescheduling)
• Classical static navigation used on-board of
  professional vehicles.

26
Conclusions RGNVSS

• Operational risks are much higher than technical
  ones.
• Most existing technical risks can be averted by
  proper drivers, operations personnel (at home
  base stations) and even clients (e.g. companies)
  training. However, relevant training services and
  tools are not common today.

27
Conclusions RGNVSS contd

• Transport operators need integrated support for
  the set up, maintenance and operation of fleet
  related ADAS. Hard to find and therefore
  expensive
• ADAS-based fleet management scheme requires large
  investments, (gt 10 vehicles reported pay-back
  period of 4-8 years). Additional state funds or
  promotion policies (i.e. tax exemption) need to
  be offered to the pioneer transport operators.

28
Preliminary guidelines for RGNVSS

• Automatic verification of transmissions between
  base station and truck should be supported.
• When the truck is out of communication range,
  both the base and the driver should be
  automatically informed, as well as when
  communication has been re-established.

29
Preliminary guidelines (RGNVSS) contd

• Training packages and manuals for the driver, the
  base station personnel and even the client should
  be offered by the companies installing ADAS on
  trucks that involve communication issues.
• Data storage and transmission length limitations
  should be imposed and explained to the users.
  Relevant field ranges should be as much as
  possible adjustable, to cater for different user
  needs.

30
Preliminary guidelines (RGNVSS) contd

• Navigation maps need to extend to more European
  areas and especially south-eastern Europe and the
  Balkans, where communication links are still
  rather poor and therefore on-board solution (i.e.
  reliable navigation maps) can mitigate technical
  risks in operation due to communication errors.

31
To be evaluated (RGNVSS)

• Define the limits and optimal content of
  transmitted messages length for a list of key
  communication functionalities.
• Optimise the exchanged messages HMI, to minimise
  the required driver / base personnel training and
  enhance systems safety.

32
To be evaluated (RGNVSS) contd

• Simulate a few complex message exchanges to
  verify the message understandability and involved
  actors interactions, using the proposed vs
  current message contents and layouts.

33
Evaluation Methodology
34
Evaluation Methodology a.o.

• Traffic Safety Assessment
• Usability and driver comfort assessment
• HMI factors and Interaction safety
• Road network efficiency and Environmental Impact
  Assessment
• Toolbox and Integration into a Framework

35
Traffic Safety Assessment, a.o.

• driver errors, various approaches
• drowsiness
• stress
• situation awareness
• vehicle measures (TTC, TLC, impact vulnerable
  road users etc)
• test vehicle confirmation (sensor tests etc)

36
Usability and driver comfort assessment, e.g.

• usability scales
• subjective driver satisfaction

37
HMI factors and Interaction safety, e.g.

• HMI Model
• ADAS Quickcheck
• general criteria
• criteria related to control (input) functions
• criteria to display (output) functions

38
Road network efficiency and Environmental Impact
Assessment

• Microscopic and macroscopic modelling for network
  efficiency
• fuel consumption
• emission levels (mass) of CO2, CO, HC, NOx and
  particulates (PM).

39
Model links
40
Toolbox

• Approx. 80 different tools
• Described on various characteristics and
  references

41
Test within ADVISORS

• HMI aspects of RGNVSS for truck drivers and
  planners

42
Conclusions

• Safety and network related problems and costs a
  problem
• Dynamic navigation function most attractive
  (comfort, travel time, fuel consumption)
• Trucks ACC (safety, comfort)

43
Conclusions contd

• EN Fusion between all data give priority to
  on-board data
• Need for dynamic database
• HMI aspects, a.o. Optimal combination of warning
  modalities
• RGNVSS Operational risks
• HMI
• Training
• High investments
• Inform both parties if communication fails
• standardisation of optimal lengths of messages
• Navigation maps to include SE Europe

44
Safety is also an important issue for freight
transport

• Project on Near accident analysis. Who wants to
  be involved??
• Please contact Marion Wiethoff at
• m.wiethoff_at_tbm.tudelft.nl