It’s not that the shadow position wrong, the vehicle is slightly lifted when it gets on the bridge, it’s probably that the collision model is inaccurate. In addition, the collision model of the excavator is much larger than the actual visible area. I hope this can be fixed as well.