diff --git a/src/Autonomous_vehicle_navigation_using_deep_learning_master/README.md b/src/Autonomous_vehicle_navigation_using_deep_learning_master/README.md
index f60838bd84..3a90822a45 100644
--- a/src/Autonomous_vehicle_navigation_using_deep_learning_master/README.md
+++ b/src/Autonomous_vehicle_navigation_using_deep_learning_master/README.md
@@ -1,17 +1,85 @@
-这是manayume的使用深度学习的自动驾驶汽车导航项目测试
-参考项目：https://github.com/varunpratap222/Autonomous-Vehicle-Navigation-Using-Deep-Learning.git
-测试环境：使用ubuntu20.04版本，conda虚拟环境，python3.7版本，安装的软件包按照requirements.txt文件安装，使用Carla0.9.13版本进行模拟
+# Autonomous Vehicle Navigation Using Deep Learning
 
+本项目基于深度学习实现自动驾驶汽车在CARLA仿真环境中的导航系统，支持自定义轨迹规划和行人动态模拟。
 
-启动流程
-## Run
-1. Run Carla Server using: `./CarlaUE4.sh`
-2. Run `config.py` file to load Town02
-3. Generate traffic either using `generate_traffic.py` or spawn pedestrians at random location along Trajectory 1 and 2 using the `pedestrians_1.py` and `pedestrians_2.py` script. Change the number of vehicles and pedestrians to spawn using `generate_traffic.py` by passing the corresponding arguments.
-4. Select an existing trajectory (Trajectory 1, Trajectory 2, Trajectory 3, Trajectory 4) or set custom trajectory using the format given in `test_everything.py` arguments.
-5. To find the initial and final locations of the custom trajectory, make use of `get_location.py` file and navigate the map using W,A,S,D, E, and Q keys. 
-6. Enter locations or select existing trajectories and run the `test_everything.py` file.
+## 快速开始
 
+### 环境要求
+- **操作系统**: Ubuntu 20.04
+- **仿真环境**: CARLA 0.9.13
+- **Python**: 3.7
+- **包管理**: Conda虚拟环境
 
+### 安装步骤
 
+1. **安装依赖包**:
+```bash
+conda create -n carla-env python=3.7
+conda activate carla-env
+pip install -r requirements.txt
+```
 
+2. **启动CARLA仿真器**:
+```bash
+./CarlaUE4.sh
+```
+
+3. **运行主程序**:
+```bash
+python main.py
+```
+
+## 项目结构
+
+```
+├── main.py                    # 主程序入口
+├── config.py                  # 配置文件
+├── get_location.py            # 位置坐标获取工具
+├── test_braking.py            # 刹车模型测试
+├── test_driving.py            # 驾驶模型测试
+├── pedestrians_1.py           # 行人生成器1
+├── pedestrians_2.py           # 行人生成器2
+└── requirements.txt           # 依赖包列表
+```
+
+## 核心功能
+
+### 1. 自定义轨迹规划
+使用 `get_location.py` 获取当前摄像头坐标，配置到 `config.py`:
+
+```python
+TRAJECTORIES = {
+    "custom_trajectory": {
+        "start": [x, y, z, yaw],  # 起点坐标和朝向
+        "end": [x, y, z],         # 终点坐标
+        "description": "自定义轨迹 - 城镇道路"
+    }
+}
+```
+
+### 2. 模型测试
+- **刹车测试**: `test_braking.py` - 验证紧急制动性能
+- **驾驶测试**: `test_driving.py` - 评估导航准确性
+
+### 3. 行人模拟
+- `pedestrians_1.py` - 随机行人生成（模式1）
+- `pedestrians_2.py` - 随机行人生成（模式2）
+## 配置说明
+
+### 关键配置文件
+`config.py` 包含所有可调整参数：
+- 轨迹起点/终点坐标
+- 深度学习模型参数
+- 仿真环境设置
+
+## 参考项目
+本项目参考自: [varunpratap222/Autonomous-Vehicle-Navigation-Using-Deep-Learning](https://github.com/varunpratap222/Autonomous-Vehicle-Navigation-Using-Deep-Learning.git)
+
+## 📝 注意事项
+1. 确保CARLA仿真器已正确启动
+2. 建议在独立的Conda环境中运行
+3. 行人模拟模块需要额外计算资源
+
+---
+
+**温馨提示**: 运行前请确认CARLA版本为0.9.13，Python版本为3.7，以避免兼容性问题。
\ No newline at end of file
diff --git a/src/Autonomous_vehicle_navigation_using_deep_learning_master/generate_traffic.py b/src/Autonomous_vehicle_navigation_using_deep_learning_master/generate_traffic.py
deleted file mode 100644
index e3717baf40..0000000000
--- a/src/Autonomous_vehicle_navigation_using_deep_learning_master/generate_traffic.py
+++ /dev/null
@@ -1,383 +0,0 @@
-#!/usr/bin/env python
-
-# Copyright (c) 2021 Computer Vision Center (CVC) at the Universitat Autonoma de
-# Barcelona (UAB).
-#
-# This work is licensed under the terms of the MIT license.
-# For a copy, see <https://opensource.org/licenses/MIT>.
-
-"""Example script to generate traffic in the simulation"""
-
-import glob
-import os
-import sys
-import time
-
-try:
-    sys.path.append(glob.glob('../carla/dist/carla-*%d.%d-%s.egg' % (
-        sys.version_info.major,
-        sys.version_info.minor,
-        'win-amd64' if os.name == 'nt' else 'linux-x86_64'))[0])
-except IndexError:
-    pass
-
-import carla
-
-from carla import VehicleLightState as vls
-
-import argparse
-import logging
-from numpy import random
-
-def get_actor_blueprints(world, filter, generation):
-    bps = world.get_blueprint_library().filter(filter)
-
-    if generation.lower() == "all":
-        return bps
-
-    # If the filter returns only one bp, we assume that this one needed
-    # and therefore, we ignore the generation
-    if len(bps) == 1:
-        return bps
-
-    try:
-        int_generation = int(generation)
-        # Check if generation is in available generations
-        if int_generation in [1, 2]:
-            bps = [x for x in bps if int(x.get_attribute('generation')) == int_generation]
-            return bps
-        else:
-            print("   Warning! Actor Generation is not valid. No actor will be spawned.")
-            return []
-    except:
-        print("   Warning! Actor Generation is not valid. No actor will be spawned.")
-        return []
-
-def main():
-    argparser = argparse.ArgumentParser(
-        description=__doc__)
-    argparser.add_argument(
-        '--host',
-        metavar='H',
-        default='127.0.0.1',
-        help='IP of the host server (default: 127.0.0.1)')
-    argparser.add_argument(
-        '-p', '--port',
-        metavar='P',
-        default=2000,
-        type=int,
-        help='TCP port to listen to (default: 2000)')
-    argparser.add_argument(
-        '-n', '--number-of-vehicles',
-        metavar='N',
-        default=35,
-       type=int,
-        help='Number of vehicles (default: 30)')
-    argparser.add_argument(
-        '-w', '--number-of-walkers',
-        metavar='W',
-        default=80,
-        type=int,
-        help='Number of walkers (default: 10)')
-    argparser.add_argument(
-        '--safe',
-        action='store_true',
-        help='Avoid spawning vehicles prone to accidents')
-    argparser.add_argument(
-        '--filterv',
-        metavar='PATTERN',
-        default='vehicle.*',
-        help='Filter vehicle model (default: "vehicle.*")')
-    argparser.add_argument(
-        '--generationv',
-        metavar='G',
-        default='1',
-        help='restrict to certain vehicle generation (values: "1","2","All" - default: "All")')
-    argparser.add_argument(
-        '--filterw',
-        metavar='PATTERN',
-        default='walker.pedestrian.*',
-        help='Filter pedestrian type (default: "walker.pedestrian.*")')
-    argparser.add_argument(
-        '--generationw',
-        metavar='G',
-        default='2',
-        help='restrict to certain pedestrian generation (values: "1","2","All" - default: "2")')
-    argparser.add_argument(
-        '--tm-port',
-        metavar='P',
-        default=8000,
-        type=int,
-        help='Port to communicate with TM (default: 8000)')
-    argparser.add_argument(
-        '--asynch',
-        action='store_true',
-        help='Activate asynchronous mode execution')
-    argparser.add_argument(
-        '--hybrid',
-        action='store_true',
-        help='Activate hybrid mode for Traffic Manager')
-    argparser.add_argument(
-        '-s', '--seed',
-        metavar='S',
-        type=int,
-        help='Set random device seed and deterministic mode for Traffic Manager')
-    argparser.add_argument(
-        '--seedw',
-        metavar='S',
-        default=0,
-        type=int,
-        help='Set the seed for pedestrians module')
-    argparser.add_argument(
-        '--car-lights-on',
-        action='store_true',
-        default=False,
-        help='Enable automatic car light management')
-    argparser.add_argument(
-        '--hero',
-        action='store_true',
-        default=False,
-        help='Set one of the vehicles as hero')
-    argparser.add_argument(
-        '--respawn',
-        action='store_true',
-        default=False,
-        help='Automatically respawn dormant vehicles (only in large maps)')
-    argparser.add_argument(
-        '--no-rendering',
-        action='store_true',
-        default=False,
-        help='Activate no rendering mode')
-
-    args = argparser.parse_args()
-
-    logging.basicConfig(format='%(levelname)s: %(message)s', level=logging.INFO)
-
-    vehicles_list = []
-    walkers_list = []
-    all_id = []
-    client = carla.Client(args.host, args.port)
-    client.set_timeout(10.0)
-    synchronous_master = False
-    random.seed(args.seed if args.seed is not None else int(time.time()))
-
-    try:
-        world = client.get_world()
-
-        traffic_manager = client.get_trafficmanager(args.tm_port)
-        traffic_manager.set_global_distance_to_leading_vehicle(2.5)
-        if args.respawn:
-            traffic_manager.set_respawn_dormant_vehicles(True)
-        if args.hybrid:
-            traffic_manager.set_hybrid_physics_mode(True)
-            traffic_manager.set_hybrid_physics_radius(70.0)
-        if args.seed is not None:
-            traffic_manager.set_random_device_seed(args.seed)
-
-        settings = world.get_settings()
-        if not args.asynch:
-            traffic_manager.set_synchronous_mode(True)
-            if not settings.synchronous_mode:
-                synchronous_master = True
-                settings.synchronous_mode = True
-                settings.fixed_delta_seconds = 0.05
-            else:
-                synchronous_master = False
-        else:
-            print("You are currently in asynchronous mode. If this is a traffic simulation, \
-            you could experience some issues. If it's not working correctly, switch to synchronous \
-            mode by using traffic_manager.set_synchronous_mode(True)")
-
-        if args.no_rendering:
-            settings.no_rendering_mode = True
-        world.apply_settings(settings)
-
-        blueprints = get_actor_blueprints(world, args.filterv, args.generationv)
-        blueprintsWalkers = get_actor_blueprints(world, args.filterw, args.generationw)
-        
-        
-
-        if args.safe:
-            blueprints = [x for x in blueprints if int(x.get_attribute('number_of_wheels')) == 4]
-            blueprints = [x for x in blueprints if not x.id.endswith('microlino')]
-            blueprints = [x for x in blueprints if not x.id.endswith('carlacola')]
-            blueprints = [x for x in blueprints if not x.id.endswith('cybertruck')]
-            blueprints = [x for x in blueprints if not x.id.endswith('t2')]
-            blueprints = [x for x in blueprints if not x.id.endswith('sprinter')]
-            blueprints = [x for x in blueprints if not x.id.endswith('firetruck')]
-            blueprints = [x for x in blueprints if not x.id.endswith('ambulance')]
-
-        blueprints = sorted(blueprints, key=lambda bp: bp.id)
-
-        spawn_points = world.get_map().get_spawn_points()
-        number_of_spawn_points = len(spawn_points)
-
-        if args.number_of_vehicles < number_of_spawn_points:
-            random.shuffle(spawn_points)
-        elif args.number_of_vehicles > number_of_spawn_points:
-            msg = 'requested %d vehicles, but could only find %d spawn points'
-            logging.warning(msg, args.number_of_vehicles, number_of_spawn_points)
-            args.number_of_vehicles = number_of_spawn_points
-
-        # @todo cannot import these directly.
-        SpawnActor = carla.command.SpawnActor
-        SetAutopilot = carla.command.SetAutopilot
-        FutureActor = carla.command.FutureActor
-
-        # --------------
-        # Spawn vehicles
-        # --------------
-        batch = []
-        hero = args.hero
-        for n, transform in enumerate(spawn_points):
-            if n >= args.number_of_vehicles:
-                break
-            blueprint = random.choice(blueprints)
-            if blueprint.has_attribute('color'):
-                color = random.choice(blueprint.get_attribute('color').recommended_values)
-                blueprint.set_attribute('color', color)
-            if blueprint.has_attribute('driver_id'):
-                driver_id = random.choice(blueprint.get_attribute('driver_id').recommended_values)
-                blueprint.set_attribute('driver_id', driver_id)
-            if hero:
-                blueprint.set_attribute('role_name', 'hero')
-                hero = False
-            else:
-                blueprint.set_attribute('role_name', 'autopilot')
-
-            # spawn the cars and set their autopilot and light state all together
-            batch.append(SpawnActor(blueprint, transform)
-                .then(SetAutopilot(FutureActor, True, traffic_manager.get_port())))
-
-        for response in client.apply_batch_sync(batch, synchronous_master):
-            if response.error:
-                logging.error(response.error)
-            else:
-                vehicles_list.append(response.actor_id)
-
-        # Set automatic vehicle lights update if specified
-        if args.car_lights_on:
-            all_vehicle_actors = world.get_actors(vehicles_list)
-            for actor in all_vehicle_actors:
-                traffic_manager.update_vehicle_lights(actor, True)
-
-        # -------------
-        # Spawn Walkers
-        # -------------
-        # some settings
-        percentagePedestriansRunning = 0.0      # how many pedestrians will run
-        percentagePedestriansCrossing = 0.0     # how many pedestrians will walk through the road
-        if args.seedw:
-            world.set_pedestrians_seed(args.seedw)
-            random.seed(args.seedw)
-        # 1. take all the random locations to spawn
-        spawn_points = []
-        for i in range(args.number_of_walkers):
-            spawn_point = carla.Transform()
-            loc = world.get_random_location_from_navigation()
-            if (loc != None):
-                spawn_point.location = loc
-                spawn_points.append(spawn_point)
-        # 2. we spawn the walker object
-        batch = []
-        walker_speed = []
-        for spawn_point in spawn_points:
-            walker_bp = random.choice(blueprintsWalkers)
-            # set as not invincible
-            if walker_bp.has_attribute('is_invincible'):
-                walker_bp.set_attribute('is_invincible', 'false')
-            # set the max speed
-            if walker_bp.has_attribute('speed'):
-                if (random.random() > percentagePedestriansRunning):
-                    # walking
-                    walker_speed.append(walker_bp.get_attribute('speed').recommended_values[1])
-                else:
-                    # running
-                    walker_speed.append(walker_bp.get_attribute('speed').recommended_values[2])
-            else:
-                print("Walker has no speed")
-                walker_speed.append(0.0)
-            batch.append(SpawnActor(walker_bp, spawn_point))
-        results = client.apply_batch_sync(batch, True)
-        walker_speed2 = []
-        for i in range(len(results)):
-            if results[i].error:
-                logging.error(results[i].error)
-            else:
-                walkers_list.append({"id": results[i].actor_id})
-                walker_speed2.append(walker_speed[i])
-        walker_speed = walker_speed2
-        # 3. we spawn the walker controller
-        batch = []
-        walker_controller_bp = world.get_blueprint_library().find('controller.ai.walker')
-        for i in range(len(walkers_list)):
-            batch.append(SpawnActor(walker_controller_bp, carla.Transform(), walkers_list[i]["id"]))
-        results = client.apply_batch_sync(batch, True)
-        for i in range(len(results)):
-            if results[i].error:
-                logging.error(results[i].error)
-            else:
-                walkers_list[i]["con"] = results[i].actor_id
-        # 4. we put together the walkers and controllers id to get the objects from their id
-        for i in range(len(walkers_list)):
-            all_id.append(walkers_list[i]["con"])
-            all_id.append(walkers_list[i]["id"])
-        all_actors = world.get_actors(all_id)
-
-        # wait for a tick to ensure client receives the last transform of the walkers we have just created
-        if args.asynch or not synchronous_master:
-            world.wait_for_tick()
-        else:
-            world.tick()
-
-        # 5. initialize each controller and set target to walk to (list is [controler, actor, controller, actor ...])
-        # set how many pedestrians can cross the road
-        world.set_pedestrians_cross_factor(percentagePedestriansCrossing)
-        for i in range(0, len(all_id), 2):
-            # start walker
-            all_actors[i].start()
-            # set walk to random point
-            all_actors[i].go_to_location(world.get_random_location_from_navigation())
-            # max speed
-            all_actors[i].set_max_speed(float(walker_speed[int(i/2)]))
-
-        print('spawned %d vehicles and %d walkers, press Ctrl+C to exit.' % (len(vehicles_list), len(walkers_list)))
-
-        # Example of how to use Traffic Manager parameters
-        traffic_manager.global_percentage_speed_difference(30.0)
-
-        while True:
-            if not args.asynch and synchronous_master:
-                world.tick()
-            else:
-                world.wait_for_tick()
-
-    finally:
-
-        if not args.asynch and synchronous_master:
-            settings = world.get_settings()
-            settings.synchronous_mode = False
-            settings.no_rendering_mode = False
-            settings.fixed_delta_seconds = None
-            world.apply_settings(settings)
-
-        print('\ndestroying %d vehicles' % len(vehicles_list))
-        client.apply_batch([carla.command.DestroyActor(x) for x in vehicles_list])
-
-        # stop walker controllers (list is [controller, actor, controller, actor ...])
-        for i in range(0, len(all_id), 2):
-            all_actors[i].stop()
-
-        print('\ndestroying %d walkers' % len(walkers_list))
-        client.apply_batch([carla.command.DestroyActor(x) for x in all_id])
-
-        time.sleep(0.5)
-
-if __name__ == '__main__':
-
-    try:
-        main()
-    except KeyboardInterrupt:
-        pass
-    finally:
-        print('\ndone.')
diff --git a/src/Autonomous_vehicle_navigation_using_deep_learning_master/main.py b/src/Autonomous_vehicle_navigation_using_deep_learning_master/main.py
index 104a8c1166..69975cea17 100644
--- a/src/Autonomous_vehicle_navigation_using_deep_learning_master/main.py
+++ b/src/Autonomous_vehicle_navigation_using_deep_learning_master/main.py
@@ -1,7 +1,6 @@
 """
 主程序入口 - 协调所有模块
 """
-
 import os
 os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
 import time
diff --git a/src/Autonomous_vehicle_navigation_using_deep_learning_master/test_everything.py b/src/Autonomous_vehicle_navigation_using_deep_learning_master/test_everything.py
deleted file mode 100644
index 6518e316f5..0000000000
--- a/src/Autonomous_vehicle_navigation_using_deep_learning_master/test_everything.py
+++ /dev/null
@@ -1,302 +0,0 @@
-import os
-os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
-import random
-from collections import deque
-import numpy as np
-import cv2
-import time
-import tensorflow as tf
-from tensorflow.keras.models import load_model
-from car_env import CarEnv, MEMORY_FRACTION
-import carla
-from carla import Transform, Location, Rotation
-
-# 轨迹定义
-trajectories = {
-    "custom_trajectory": {
-        "start": [-8.77956485748291,140.2951202392578,2.0014660358428955, 0], 
-        "end": [74.17852020263672,-56.52183151245117,0.18172569572925568],
-        "description": "自定义轨迹"
-    }
-}
-
-SELECTED_TRAJECTORY = "custom_trajectory"
-
-def get_selected_trajectory():
-    """获取选定的轨迹"""
-    if SELECTED_TRAJECTORY in trajectories:
-        trajectory = trajectories[SELECTED_TRAJECTORY]
-        print(f"✅ 使用轨迹: {SELECTED_TRAJECTORY}")
-        print(f"  描述: {trajectory['description']}")
-        print(f"  起点: {trajectory['start']}")
-        print(f"  终点: {trajectory['end']}")
-        return trajectory
-    else:
-        print(f"❌ 轨迹 '{SELECTED_TRAJECTORY}' 不存在")
-        return None
-
-def safe_load_model(model_path):
-    """安全加载模型"""
-    try:
-        if not os.path.exists(model_path):
-            print(f"❌ 模型文件不存在: {model_path}")
-            return None
-            
-        model = load_model(model_path)
-        print(f"✅ 成功加载模型: {model_path}")
-        return model
-        
-    except Exception as e:
-        print(f"❌ 加载模型失败: {e}")
-        return None
-
-def setup_tensorflow():
-    """设置 TensorFlow 配置"""
-    os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
-    print(f"TensorFlow 版本: {tf.__version__}")
-    
-    # GPU 配置
-    gpus = tf.config.list_physical_devices('GPU')
-    if gpus:
-        try:
-            for gpu in gpus:
-                tf.config.experimental.set_memory_growth(gpu, True)
-            print(f"✅ 找到 {len(gpus)} 个GPU，已启用内存增长")
-        except RuntimeError as e:
-            print(f"⚠️ GPU设置错误: {e}")
-            os.environ['CUDA_VISIBLE_DEVICES'] = '-1'
-            print("使用CPU运行")
-    else:
-        print("ℹ️ 未找到GPU，使用CPU运行")
-
-def set_spectator_to_vehicle(world, vehicle):
-    """设置观察者视角 - 车辆正后方跟随"""
-    try:
-        spectator = world.get_spectator()
-        transform = vehicle.get_transform()
-        
-        # 计算车辆后方的位置
-        # 使用车辆的旋转来确定方向
-        rotation = transform.rotation
-        yaw = np.radians(rotation.yaw)
-        
-        # 在车辆后方一定距离（例如8米），高度3米
-        distance_behind = 8.0
-        height = 3.0
-        
-        # 计算后方位置（与车辆朝向相反的方向）
-        behind_x = transform.location.x - distance_behind * np.cos(yaw)
-        behind_y = transform.location.y - distance_behind * np.sin(yaw)
-        
-        # 设置观察者在车辆正后方，稍微高一点
-        spectator.set_transform(Transform(
-            Location(x=behind_x, y=behind_y, z=transform.location.z + height),
-            Rotation(pitch=-15, yaw=rotation.yaw)  # 与车辆相同的水平朝向
-        ))
-        
-    except Exception as e:
-        print(f"⚠️ 设置视角时出错: {e}")
-
-def preprocess_state_for_prediction(state_data, model_type="braking"):
-    """预处理状态数据用于模型预测"""
-    try:
-        if model_type == "braking":
-            state_array = np.array(state_data[:2])
-        else:
-            state_array = np.array(state_data[2:])
-        
-        if len(state_array.shape) == 1:
-            state_array = state_array.reshape(1, -1)
-        
-        return state_array
-    except Exception as e:
-        print(f"状态预处理错误: {e}")
-        return np.array([[0, 0]])
-
-def debug_vehicle_state(vehicle):
-    """调试车辆状态"""
-    if vehicle is None:
-        print("❌ 车辆为 None")
-        return
-    
-    try:
-        transform = vehicle.get_transform()
-        velocity = vehicle.get_velocity()
-        print(f"📍 车辆位置: ({transform.location.x:.2f}, {transform.location.y:.2f}, {transform.location.z:.2f})")
-        print(f"🧭 车辆朝向: {transform.rotation.yaw:.2f}°")
-        print(f"🚀 车辆速度: {np.sqrt(velocity.x**2 + velocity.y**2 + velocity.z**2):.2f} m/s")
-    except Exception as e:
-        print(f"❌ 获取车辆状态失败: {e}")
-
-def main():
-    # 设置 TensorFlow
-    setup_tensorflow()
-    
-    # 获取选定的轨迹
-    trajectory = get_selected_trajectory()
-    if trajectory is None:
-        print("❌ 无法获取轨迹，退出程序")
-        return
-        
-    start_location = trajectory["start"]
-    end_location = trajectory["end"]
-    
-    # 加载模型
-    print("\n" + "="*50)
-    print("加载自动驾驶模型")
-    print("="*50)
-    
-    MODEL_PATH = "models/Braking___282.model"
-    MODEL_PATH2 = "models/Driving__6030.model"
-    
-    model = safe_load_model(MODEL_PATH)
-    model2 = safe_load_model(MODEL_PATH2)
-    
-    if model is None or model2 is None:
-        print("❌ 模型加载失败，退出程序")
-        return
-    
-    # 创建环境
-    print("\n初始化CARLA环境...")
-    try:
-        env = CarEnv(start_location, end_location)
-        world = env.client.get_world()
-        
-        # 设置仿真设置
-        settings = world.get_settings()
-        settings.synchronous_mode = False
-        settings.fixed_delta_seconds = 0.05
-        world.apply_settings(settings)
-        
-    except Exception as e:
-        print(f"❌ 初始化环境失败: {e}")
-        return
-    
-    # 主循环
-    fps_counter = deque(maxlen=60)
-    EPISODES = 2
-    
-    for episode in range(EPISODES):
-        print(f'\n{"="*50}')
-        print(f'开始 Episode {episode + 1}/{EPISODES}')
-        print(f'{"="*50}')
-        
-        # 重置环境 - 这会生成车辆
-        try:
-            print("重置环境...")
-            current_state = env.reset()
-            print(f"初始状态: {current_state}")
-        except Exception as e:
-            print(f"❌ 环境重置失败: {e}")
-            continue
-        
-        # 从环境中获取车辆
-        ego_vehicle = env.vehicle
-        
-        if ego_vehicle is None:
-            print("❌ 环境中没有车辆，跳过此episode")
-            continue
-        
-        # 调试车辆状态
-        debug_vehicle_state(ego_vehicle)
-        
-        # 设置观察者视角
-        set_spectator_to_vehicle(world, ego_vehicle)
-        
-        done = False
-        step_count = 0
-        max_steps = 1000
-        
-        while not done and step_count < max_steps:
-            step_count += 1
-            step_start = time.time()
-            
-            # 定期更新视角
-            if step_count%5==0:
-                set_spectator_to_vehicle(world, ego_vehicle)
-            
-            # 动作预测
-            action = 0
-            try:
-                # 检查交通灯
-                if hasattr(env, 'vehicle') and env.vehicle and env.vehicle.is_at_traffic_light():
-                    traffic_light = env.vehicle.get_traffic_light()
-                    if traffic_light and traffic_light.get_state() == carla.TrafficLightState.Red:
-                        print("🚦 红灯 - 停车")
-                        action = 0
-                    else:
-                        # 使用模型预测
-                        state_array = preprocess_state_for_prediction(current_state, "braking")
-                        qs = model.predict(state_array, verbose=0)[0]
-                        action = np.argmax(qs)
-                        
-                        if action == 1:  # 安全时才使用驾驶模型
-                            state_array2 = preprocess_state_for_prediction(current_state, "driving")
-                            qs2 = model2.predict(state_array2, verbose=0)[0]
-                            action = np.argmax(qs2) + 1
-                else:
-                    # 正常情况下的决策
-                    state_array = preprocess_state_for_prediction(current_state, "braking")
-                    qs = model.predict(state_array, verbose=0)[0]
-                    action = np.argmax(qs)
-                    
-                    if action == 1:
-                        state_array2 = preprocess_state_for_prediction(current_state, "driving")
-                        qs2 = model2.predict(state_array2, verbose=0)[0]
-                        action = np.argmax(qs2) + 1
-                        
-            except Exception as e:
-                print(f"❌ 预测错误: {e}")
-                action = 0
-            
-            # 执行动作
-            try:
-                new_state, reward, done, waypoint = env.step(action, current_state)
-                current_state = new_state
-                
-                # 显示额外信息
-                if step_count % 10 == 0:
-                    print(f"步骤 {step_count}, 奖励: {reward}, 完成: {done}")
-                
-            except Exception as e:
-                print(f"❌ 环境步骤错误: {e}")
-                done = True
-            
-            # 计算FPS
-            frame_time = time.time() - step_start
-            fps_counter.append(frame_time)
-            current_fps = len(fps_counter) / sum(fps_counter) if fps_counter else 0
-            
-            # 显示动作名称
-            action_names = ["刹车", "直行", "左转", "右转", "微左", "微右"]
-            action_name = action_names[action] if action < len(action_names) else str(action)
-            
-            print(f'Step: {step_count:>3d} | FPS: {current_fps:>4.1f} | Action: {action_name}')
-            
-            if done:
-                print(f"Episode {episode + 1} 完成，步数: {step_count}")
-                break
-        
-        if step_count >= max_steps:
-            print(f"Episode {episode + 1} 达到最大步数限制")
-        
-        # 等待一段时间再开始下一个episode
-        print(f"等待下一个episode...")
-        time.sleep(2.0)
-    
-    # 最终清理
-    print("\n" + "="*50)
-    print("所有episodes完成!")
-    print("="*50)
-    
-    print("清理资源...")
-    try:
-        # 环境会在重置时自动清理车辆
-        cv2.destroyAllWindows()
-    except:
-        pass
-    
-    print("程序结束")
-
-if __name__ == '__main__':
-    main()
\ No newline at end of file