Informal Document: ACSF Rev.1

Slides:



Advertisements
Similar presentations
Grading Scheme Midterm—20% Final Exam—25% Teacher Evaluation—55% Presentation (10%) Class Participation (10%) Safety and Health Behavior (25%) Class Performance.
Advertisements

Parallel Parking Identify 1½ Car Lengths of Available Space
The necessity of New Regulations for New Technologies regarding R79 Japan September / 2012 Informal document GRRF (73rd GRRF, September 2012,
ACSF Informal Group Industry proposals 1 st Meeting of ACSF informal group April 29 and 30, 2015 in Bonn 1 Informal Document ACSF
1 Different use cases for ACSF Prepared from the experts of OICA and CLEPA Informal Document ACSF
3rd ACSF meeting Munich, September 2-3 Industry proposals for ACSF status definition and HMI Informal Document: ACSF Submitted by the experts from.
Outline of Definition of Automated Driving Technology Document No. ITS/AD (5th ITS/AD, 24 June 2015, agenda item 3-2) Submitted by Japan.
Presentation for Document ACSF-03-03_rev1 Oliver Kloeckner September rd meeting of the IG ASCF Munich, Airport Informal Document.
Damage Mitigation Braking System
IntelliDriveSM Update
Ministry of Land, Infrastructure, Transport and Tourism Possible Example of Split of Regulation ~ Split of R79 and The New Regulation of Driving Assistance.
Difference for "Automatically commanded steering function", "Corrective steering function" and " Autonomous Steering System " Informal Document: ACSF
1 ACSF Test Procedure Draft proposal – For discussion OICA and CLEPA proposal for the IG Group ACSF Tokyo, 2015, June Informal Document ACSF
IHRA-ITS UN-ECE WP.29 ITS Informal Group Geneva, March, 2011 Design Principles for Advanced Driver Assistance Systems: Keeping Drivers In-the-Loop Transmitted.
Remote Control Parking (RCP)
Identification of regulatory needs for ACSF Oliver Kloeckner 16-17th June nd meeting of the IG ASCF Tokyo – Jasic Office Informal Document.
Protective Braking for ACSF Informal Document: ACSF
Minimum Risk Manoeuvres (MRM)
Safety Distances and Object Classifications for ACSF Informal Document: ACSF
Results of the Study on ACSF Transition Time Informal Document: ACSF National Traffic Safety and Environment Laboratory, Japan 4th Meeting of ACSF.
5 th ACSF meeting French views Bonn January 2016 Informal Document - ACSF Submitted by the expert of France.
ACSF - Traffic Jam Assist on all roads OICA and CLEPA proposal for the IG Group ACSF Tokyo, 2015, June Informal Document ACSF
Sample Workbook for Crash Avoidance Technologies Focus Groups.
1 6th ACSF meeting Tokyo, April 2016 Requirements for “Sensor view” & Environment monitoring version 1.0 Transmitted by the Experts of OICA and CLEPA.
Sample Workbook for Crash Avoidance Technologies Focus Groups.
Transmitted by the Experts of TRL (EC)
Informal document GRRF-84-32
Informal Document: ACSF-06-16
Unit 5 VEHICLE HANDLING SAFE VEHICLE CONTROL
On-board Technologies
ADVANCED DRIVER ASSISTANCE SYSTEMS
Discussion paper – Major Issues
7th ACSF meeting London, June 28-30, 2016
Presentation of ACSF C tests
Informal Working Group on ACSF
ACSF-C2 2-actions system
Automatic Emergency Braking Systems (AEBS)
Timing to be activated the hazard lights
ACSF-C2 2-actions system
Working Principle of Blind Spot Technology in Car
The Safety Technologies to Consider While Buying a New Car.
Comparison of Cat.C HMI solution and vehicle without Cat.C
TUGS Jason Higuchi && Julia Yefimenko && Raudel mayorga
Industry Homework from AEB 02
Submitted by the experts of OICA
Status of the Informal Working Group on ACSF
Proposals from the Informal Working Group on AEBS
Informal document GRSG
ACSF C tests Prepared by JAPAN 15 November, 2017
Quintessences Proposal for Category C of Germany and Japan
2018 Summit of the National Association of State Motorcycle Safety
ACSF B1+C functional description
On-board Technologies
Reason for performance difference between LVW and GVW
Proposals from the Informal Working Group on AEBS
Comparison of Cat.C HMI solution and vehicle without Cat.C
Hands-off detection warning time for B1-systems
Informal Document: ACSF-10-08
Definition of aysmax Interpretation 1
Safety considerations on Emergency Manoeuver
Maximum allowable Override Force
ACSF - Category C1 Definition of Sensor range and safety distances
ACSF B1+C functional description
Emergency Steering Function
Interpretation of CSF warnings #2
ACSF B2 SAE Level 2 and/or Level 3
6th ACSF meeting Tokyo, April 2016
Status of the Informal Working Group on ACSF
Regulation ECE R79-03 ACSF C 2-Step HMI
Automated Lane Keeping Systems
Presentation transcript:

Informal Document: ACSF-10-08-Rev.1 Content Submitted by the experts of OICA and CLEPA Informal Document: ACSF-10-08-Rev.1 ESF definition Description of use cases Taxonomy of collision avoidance functions Description of functions Tests

ESF Definition 2.3.4.3 “Emergency Steering Function (ESF)" means a control function within an electronic control system whereby, for a limited duration, changes to the steering angle of one or more wheels may result from the automatic evaluation of signals initiated on-board the vehicle, in order to assist the driver in avoiding [or mitigating] a collision with: another vehicle driving in an adjacent lane, drifting towards the path of the subject vehicle and/or, into which path the subject vehicle is drifting and/or, into which lane the driver initiates a lane change manoeuver. an obstacle obstructing the path of the subject vehicle or when the obstruction of the subject vehicle’s path is deemed imminent.

Function is expected to intervene around this point Use Cases i.a Function is expected to intervene around this point i.b i.c ii.

Taxonomy of collision avoidance functions Avoidance Manoeuver manually performed by driver (by turning steering wheel) with system support Avoidance Manoeuver automatically initiated by system Within or outside the lane (driver’s decision) Within the lane Outside the lane (system’s decision) Use cases Functions # ii. i.a i.b i.c ii. i.a ii. 1 2 3 4 5 6 7 On the market - Close to market / IN SCOPE of discussions - Regulatory step 1 Future / OUT OF SCOPE Regulatory step 2

Rationale #1 CSF CSF LKAS Stability… Collision avoid. Lane departure avoid. ACSF B1 LKAS Objectives Avoid misuse  hands-off detection Minimum performance and tests, to ensure a “safe” coupling with ACSF C/D CSF LKAS Stability… Collision avoid. Lane departure avoid. Objectives: Avoid misuse  acoustic warning if too long or too frequent intervention ESF Collision avoid. Objectives Avoid misuse  acoustic warning at every intervention (unless avoidance performed by the driver) Minimum performance and tests Industry explored the possibility to define ESF performance and tests (with e.g. TTC) but this revealed to be impractical, while not strictly required to reach the objectives.

ESF Tests Since Osaka ACSF-09 meeting, industry explored the possibility to define performance and tests for ESF. This approach led us to a dead-end, and revealed the definition of performance and tests are not required to achieve the main objective, which is to avoid misusing ESF as a B1 system. Industry proposal is to follow the same approach as for CSF (while making it more stringent): Acoustic and optical warning at every avoidance manoeuver automatically initiated by system Acoustic and/or optical warning when avoidance manoeuver manually performed by driver Describes the conditions for collision risk detection and demonstrates them 50N max overriding test (to be tested) Rationales in next slides

Use Case i.a “Another vehicle drifts towards the subject vehicle” Function type 1 2 3 4 5 6 7 2 Description: The function automatically intervenes by steering when detecting another vehicle is drifting towards its path. Function : On the market Avoidance within the lane. Forward and side looking sensors. Detection of lane marking / boundaries. The system is able to work on a road without lane marking (provided road boundaries are detected). Warning during intervention Blind spot detection may separately provide a warning prior ESF intervention Future evolution Function able to cross lane markings. 2 2 6 No intervention Intervention

Use Case i.b “The subject vehicle drifts towards the adjacent lane” Function type 1 2 3 4 5 6 7 Description: On the market The function automatically intervenes by steering when detecting: a vehicle in the blind spot or an oncoming vehicle at higher speed. The manoeuver ends: after the subject vehicle is back in the “center” of the original lane, or after “stopping the drift”, as specified by VM Avoidance within the lane. Forward and side looking sensors. Detection of lane marking / boundaries. The system is able to work on a road without lane marking (provided road boundaries are detected). Warning during intervention Blind spot detection may separately provide a warning prior ESF intervention No risk of collision  No intervention No intervention Intervention

Use Case i.c “The driver of the subject vehicle performs a lane change” Function type 1 2 3 4 5 6 7 Description: On the market The function automatically intervenes by steering when detecting: a vehicle in the blind spot or an oncoming vehicle at higher speed. The manoeuver ends: after the subject vehicle is back in the “center” of the original lane, or after aborting the LC Avoidance within the lane. Forward and side looking sensors. Detection of lane marking / boundaries. The system is able to work on a road without lane marking (provided road boundaries are detected). Warning during intervention Blind spot detection may separately provide a warning prior ESF intervention No intervention Intervention

Use Case ii. Function type “An obstacle obstructs the path” “Obstruction of the path is imminent” 1 2 3 4 5 6 7 Description: On the market. The driver performs the evasive manoeuver The system assists the driver to avoid collision The driver decides to cross the lane marking, not the system Lane markings not required Only forward looking sensors needed The system does not check side or rear traffic more details on next slide No intervention Intervention

Use Case ii. Function type “An obstacle obstructs the path” “Obstruction of the path is imminent” 1 2 3 4 5 6 7 Description: When the driver performs an evasive manoeuvre (red path) by turning the steering wheel, the system provides assistance by adding precisely calculated steering torque (i.e. amplifying driver’s torque) to support the movement of the steering wheel and avoid the collision. Given the added torque is limited and the driver is hands-on, the driver is free to follow the steering recommendation given by the system or to override the system recommendation. If the driver does not perform any evasive manoeuver, the function does not intervene. If the driver performs a second steering actuation to the right (green path), the system will assist the same way as during the red actuation, to follow the intended path. If the driver does not performs a second steering actuation to the right, the system may provide a steering recommendation to the right, which again the driver is free to follow or not. This is very similar to what a CSF function of type (c) is doing to prevent lane departure.

Use Case ii. Function type “An obstacle obstructs the path” “Obstruction of the path is imminent” 1 2 3 4 5 6 7 Description: On the market / close to market. The system performs automatic avoidance within the lane Only forward looking sensors required The system does not check side or rear traffic No intervention Intervention

Use Case ii. Function type “An obstacle obstructs the path” “Obstruction of the path is imminent” 1 2 3 4 5 6 7 Description: Future evolution, out of scope Lane markings not required Forward, side and rearward looking sensors required. The system checks if there is enough space to perform the avoidance manoeuver No intervention Intervention

Rationale #2 Time to Collision (TTC) - AEBS vs ESF AEBS for LCV ESF Assumptions: 3s TTC at 80km/h (= 67m) Stationary target 1 vehicle is moving in 1 dimension * 2 vehicles are moving in 2 dimensions 67m 67m Drift rate = [0.5m/s] (note: in R130, rate of departure between 0.1 and 0.8 m/s) 80km/h 67m Test complexity is not Comparable !! 80km/h 3 m * With a moving target at 67km/h, 2 vehicles are moving but the distance become 11 m…

2 vehicles are moving in 2 dimensions Conclusions ESF Such a test is impractical: TTC not measurable, due to high tolerances on drift rate, leading to variation in the collision point. In many cases, a TTC is not relevant for collision risk detection; there are other potential criteria (safety distance, speed, vehicle type, road conditions, use case etc.). Industry proposal is to stick to the CSF approach and provide a warning at every intervention of ESF. In addition, the manufacturer should describe the conditions for collision risk detection and demonstrate them with documentation, test results, virtual or physical tests etc. 50N max overriding test (to be tested) This will ensure no misuse and easy overriding. 2 vehicles are moving in 2 dimensions Drift rate = [0.5m/s] (note: in R130, rate of departure between 0.1 and 0.8 m/s) 67m 80km/h 3 m