You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, vehicles only enact their own speed-advice with the delay of 1 action step. This often fails to create the necessary gaps in a timely manner.
The problem is especially pronounced if action-steps of interacting vehicle do not happen in the same simulation step. Coordinated gap-creation works badly with out-of-sync speed adjustments because vehicles react to the state that is observed in their own action step and request cooperative speed adjustments accordingly. By the time the other vehicle can make use of the speed request, it's own state has changed and the advice is no longer ideal.
Non-synchronized action-steps easily appear due to sub-second depart delay or explicit sub-second departure times.
One possible solution would be to increase the frequency of patchSpeed and lane change attempts (e.g. to the step-length) once a maneuver failed. This would fit the assumption that drivers enact a planned speed-change and lane-change maneuver rather than react with delay to unforeseen developments.
The text was updated successfully, but these errors were encountered:
Currently, vehicles only enact their own speed-advice with the delay of 1 action step. This often fails to create the necessary gaps in a timely manner.
The problem is especially pronounced if action-steps of interacting vehicle do not happen in the same simulation step. Coordinated gap-creation works badly with out-of-sync speed adjustments because vehicles react to the state that is observed in their own action step and request cooperative speed adjustments accordingly. By the time the other vehicle can make use of the speed request, it's own state has changed and the advice is no longer ideal.
Non-synchronized action-steps easily appear due to sub-second depart delay or explicit sub-second departure times.
One possible solution would be to increase the frequency of patchSpeed and lane change attempts (e.g. to the step-length) once a maneuver failed. This would fit the assumption that drivers enact a planned speed-change and lane-change maneuver rather than react with delay to unforeseen developments.
The text was updated successfully, but these errors were encountered: