Time To React (TTR)#
Description#
The TTR metric [Hillenbrand2006] [Tamke2011] approximates the latest time until a reaction is required by aggregating the maximum TTM metric over a predefined set of maneuvers \(M\), i.e.
For example, as a set of maneuvers, one might select \(M = \{\)‘brake’, ‘steer’, ‘kickdown’\(\}\).
Properties#
Run-time capability#
Yes
Target values#
Depends on maneuver set \(M\)
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, set of maneuvers \(M\) and their MMs
Output scale#
\(\{-\infty\} \cup [0,\infty]\), time (s), ratio scale
Reliability#
High, under the assumption that collisions can be reliably predicted
Validity#
High, claimed to be ‘an adequate metric to assess the criticality […] since it directly relates to the driver’s possible actions’ [Hillenbrand2006], but also dependent on maneuvers and collision prediction
Sensitivity#
Comparable to TTM, as, due to selection of the maneuver with the maximal TTM, the same reasoning applies
Specificity#
Highly increased compared to TTM due to consideration of multiple maneuvers
Prediction model#
Time window#
Unbound, but usefulness depends on DMM
Time mode#
Linear time for \(|M|=1\), branching time for \(|M|>1\)