Sim-to-Real Reinforcement Learning for Autonomous Driving Using Pseudosegmentation Labeling and Dynamic Calibration

Figure 7

Results of linear regression between steering and ∆θ, which is the change of heading angle.