Hi DBNet team,
thanks for the great dataset and here comes some question.
According to common sense and Figure 4 in your paper, the wheel angle should be between -90 and 90. However, the angle of the behavior.csv in the dataset are distributed between -180 and 180. It is difficult to predict such a wide range, could you tell me the reason for that?
And the lstm results are far from the paper.
Cheers!
Hi DBNet team,
thanks for the great dataset and here comes some question.
According to common sense and Figure 4 in your paper, the wheel angle should be between -90 and 90. However, the angle of the behavior.csv in the dataset are distributed between -180 and 180. It is difficult to predict such a wide range, could you tell me the reason for that?
And the lstm results are far from the paper.
Cheers!