优化 PPO 清扫策略

agent_ppo/conf/train_env_conf.toml

@@ -6,7 +6,7 @@ map = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
 # Whether to randomly select maps. Boolean. 
 # true = randomly pick one from configured maps per episode, false = used sequentially.
 # 是否随机抽取地图。布尔值。true表示每局从配置的地图中随机抽取一张，false表示按顺序抽取地图训练。
-map_random = false
+map_random = true
 
 # Number of official robots. Range: 1~4 (integer). 
 # In each round, official robots will be randomly generated on the road according to the configured.
@@ -23,4 +23,4 @@ max_step = 1000
 
 # Maximum battery. The battery level when fully charged. Range: 100~999.
 # 最大电量。满电状态下的电量。可配置范围100～999。
-battery_max = 200
+battery_max = 200