Multiagent Cooperative Learning Strategies for Pursuit-Evasion Games

Table 2

Average number of capture steps for different movement types.

	Grid size	Clockwise movement; (steps)	Random clockwise and counterclockwise movement; (steps)
Hysteretic Q-Learning	12 * 12	56	69
RL-Fusing	12 * 12	48	56
The proposed approach	12 * 12	26	32