driving scenarios on-going

Euro NCAP driving scenarios

Adaptive Cruise Control(ACC)

ACC is tested in an extended version of AEB. these systems are designed to automatically adapt the speed when approaching a slower moving vehicle or noe that is braking. notification, not all systems perform well with stationary obstacles.

a slow moving vehicle test
a stationary obstacle(e.g. vehicle) test

autonomous emergency braking(AEB)

AEB pedestrain

an adult pedestrain running from the driver’s side of the vehicle
an adult walk from the passenger’s side
a child runs from between parked cars on the passenger’s side of the car
pedestrain walking in the same direction as the vehicle algnedwith the centre of vehicle
pedestrain walking in the same direction as the vehicle offset to one side

AEB cyclist

cyclist is crossing the vehicle’s path
cyclist is travelling in the same direction

Euro NCAP gives greatest reward if a collision is completely avoided; in some case, if AEB is unable to stop the vehicle completely, still give some points.

lane keeping assist(LKA)

lane centering, a good ADS should continue to support the driver during the manoeuvre and will not resist the driver or deactivate.
cut-in test, a npc from adjacent lane merges into the current lane
cut-off test, a npc leaves the current lane abruptly to avoid a stopped vehicle ahead
swerve around a small obstacle test
speed assistance test, to use map data and/or data from sensors to identify the local speed limit

NHTSA driving scenarios

a previous blog design scenarios in ADS simulation is based on NHTSA.

forward collision prevention

Forward collision warning
AEB
pedestrain AEB
Adaptive lighting

backing up & parking

rear automatic braking
backup camera
rear cross traffic alert

lane & side assist

lane departure warning
lane keeping assist
blind spot detection
lane centering assist

maintaing safe distance

traffic jam assist
highway pilot
ACC

Matlab Driving Scenario samples

AEB (for vulnerable road users)

AEB_Bicyclist_Longitudinal_25width, at collision time, the bicycle is 25% of the way across the width of the ego vehicle.
AEB_CCRb_2_initialGap_12m, a car-to-car rear braking(CCRb) scenario where the ego rear hit a braking agent. the deceleration is 2 m/s^2 with initial gap 12m
AEB_CCRm_50overlap, a car-to-car rear moving(CCRm) scenario, where the ego rear hits a moving vehicle. at collisio ntime, the ego overlaps with 50% of the width of the moving vehicle

AEB_CCRs_-75overlap, a car-to-car stationary(CCRs) sceanrio, where the ego rear hits a stationary vehicle. at collision time, the ego overlaps with 75% of the width of the stationary vehicle. when the ego is to the left of the other vehicle, the percent overlap is negative

AEB_Pedestrain_Farside_50width, the ego collides with a pedestrain who is traveling from the left side(far side, assuming vehicle drive on right side of the road). at collision time, the pedestrain is 50% of the way across the width of the ego
AEB_PedestrainChild_Nearside_50width, the ego collides with a pedestrain who is traveling from the right side(near side), at collision time, the pedestrain is 50% of the way across the width of the ego

Emergency Lane Keeping(ELK)

ELK_FasterOvertakingVeh_Intent_Vlat_0.5, ego intentionally changes lanes but collides with a faster overtaking vehicle, the ego travels at a lateral velocity of 0.5m/s
ELK_OncomingVeh_Vlat_0.3, ego unintentionally changes lanes and collides with an oncoming vehicle, with ego’s lateral velocity of 0.3m/s
ELK_OvertakingVeh_Unintent_Vlat_0.3, ego unintentionally changes lanes, overtake a vehicle in the other lane and collides. the ego travles at a lateral velocity at 0.3m/s
ELK_RoadEdge_NoBndry_Vlat_0.2, ego unintentionally changes lanes and ends up to hte road edge, with lateral velocity at 0.2m/s

LKA

LKA_DashedLine_Solid_Left_Vlat_0.5
LKA_DashedLine_Unmarked_Right_Vlat_0.5
LKA_RoadEdge_NoBndry_Vlat_0.5
LKA_RoadEdge_NoMarkings_Vlat_0.5
LKA_SolidLine_Dashed_Left_Vlat_0.5
LKA_SolidLine_Unmarked_Right_Vlat_0.5

open source libs

scenarios in L2 vs L3+

scenarios in l2 is more like point in the whole driving manuevor, at a certain scenario/point, how the ADAS system response. while L3+ is a continously scenario, along each point during driving is part of the scenario, and inside which all kinds of l2 case scenario can happen. it does make sense to integrate l2 scenario planning response to l3+, which makes l3+ system more strict.