Skip to content

Why the reward is to set like that? #2

@zengjie617789

Description

@zengjie617789

i donot know what the reason of the setting about reward, why does the distance reward divide the sqrt(200), and the anglereward as well ,anyone who can explain about it? thank you ~

distanceReward=-distance/np.sqrt(200)

		if abs(angle-euler[2])>np.pi:
			angleReward=-abs(-angle-euler[2])/(np.pi)
		else:
			angleReward=-abs(angle-euler[2])/(np.pi)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions