gym/gym/envs/mujoco/humanoid-new.py [67:80]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
                done = False
        elif self.realgoal == 1:
            pos_after = mass_center(self.model)
            alive_bonus = 5.0
            data = self.model.data
            lin_vel_cost = 0.25 * (pos_after - pos_before) / self.model.opt.timestep
            quad_ctrl_cost = 0.1 * np.square(data.ctrl).sum()
            quad_impact_cost = .5e-6 * np.square(data.cfrc_ext).sum()
            quad_impact_cost = min(quad_impact_cost, 10)
            reward = 0 - quad_ctrl_cost - quad_impact_cost
            qpos = self.model.data.qpos
            if not bool((qpos[2] < 1.0)):
                reward += alive_bonus + lin_vel_cost
            done = bool((qpos[2] < 1.0))
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



gym/gym/envs/mujoco/humanoid.py [73:86]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
            done = False
        elif self.realgoal == 1:
            pos_after = mass_center(self.model)
            alive_bonus = 5.0
            data = self.model.data
            lin_vel_cost = 0.25 * (pos_after - pos_before) / self.model.opt.timestep
            quad_ctrl_cost = 0.1 * np.square(data.ctrl).sum()
            quad_impact_cost = .5e-6 * np.square(data.cfrc_ext).sum()
            quad_impact_cost = min(quad_impact_cost, 10)
            reward = 0 - quad_ctrl_cost - quad_impact_cost
            qpos = self.model.data.qpos
            if not bool((qpos[2] < 1.0)):
                reward += alive_bonus + lin_vel_cost
            done = bool((qpos[2] < 1.0))
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -