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

Table 4

Student’s t-test results for comparing PPO2 with other RL algorithms.

Method	T	value	Conclusion
SAC	6.94	1.7276e − 06	Reject H0
A2C	11.40	1.1552e − 09	Reject H0
TD3	3.03	0.007248	Reject H0