Time To Steer (TTS)#

Description#

Please refer to the TTM.

Properties#

Run-time capability#

Yes

Target values#

0.47 s [Junietz2018a] (threshold for critical), 1 s [Huber2020] (ADF testing)

Subject type#

Optimal for road vehicles (automated and human), sub-optimal for VRUs

Scenario type#

Overlapping predicted trajectories for a significant time span in the scenario

Inputs#

Static/dynamic objects and their state (pose, shape, etc.) at time t, MM for maneuver ‘steer’

Output scale#

\(\{-\infty\} \cup [0,\infty]\), time (s), ratio scale

Reliability#

High, under the assumption that collisions can be reliably predicted

Validity#

High for two-actor scenes, medium for more actors when TTS is evaluated under a fixed perspective, but can be increased by aggregating over all actors as at least for one of them, the TTS will be valid

Sensitivity#

High for non-humans, as steering is often optimal for threat mitigation than braking [Adams1994]; medium for humans as steering maneuvers are seldomly preferred [Adams1994]

Specificity#

High under the assumption that the DMM has a low ratio of spurious collision among all predicted collision