I will try and explain this as best as I can.
I use RichAI and RVO.
I set RichAI.destination. If reachedDestination becomes true isStopped is set to true.
I then set isStopped = false. I then set the same destination again (the character has a destination which is set over and over).
After this, reachedDestination is then called from MovementUpdateInternal, at RichAI.cs:423
If you set a breakpoint on in reachedDestination you now see that distanceToSteeringTarget and steeringTarget does not match. distanceToSteeringTarget is the distance to the target but steeringTarget is set close to the current position.
The result is that we now think that we are 2 times further away from the target, and we also see that the character moves too close to the target (about half the endReachedDistance).
I don’t know the code well enough, but it might be that this (at RichAI.cs:433) is the cause:
steeringTarget = simulatedPosition;
I solved this by not setting isStopped = true, and I see that is not needed. I still think there is a bug here.