def main(): client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town02') weather = carla.WeatherParameters( cloudiness=30.0, precipitation=30.0, sun_altitude_angle=50.0 ) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.bmw.grandtourer') # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) origin_map.resize((2000,2000)).show() agent = BasicAgent(vehicle, target_speed=MAX_SPEED) destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, origin_map) """ while True: if close2dest(vehicle, destination): destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, origin_map) if vehicle.is_at_traffic_light(): traffic_light = vehicle.get_traffic_light() if traffic_light.get_state() == carla.TrafficLightState.Red: traffic_light.set_state(carla.TrafficLightState.Green) control = agent.run_step() control.manual_gear_shift = False vehicle.apply_control(control) nav = get_nav(vehicle, plan_map) cv2.imshow('Nav', nav) cv2.waitKey(16) """ cv2.destroyAllWindows() vehicle.destroy()
def main(): global global_nav, global_vel, start_control, global_plan_map, global_vehicle, global_cost_map, global_transform, max_steer_angle, global_a, draw_cost_map, state0 client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters( cloudiness=random.randint(0,10), precipitation=0, sun_altitude_angle=random.randint(50,90) ) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_vehicle = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() max_steer_angle = np.deg2rad(physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera':{ 'transform':carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback':image_callback, }, 'camera:view':{ #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform':carla.Transform(carla.Location(x=-3.0, y=0.0, z=6.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback':view_image_callback, }, 'lidar':{ 'transform':carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback':lidar_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() #spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) agent = BasicAgent(vehicle, target_speed=MAX_SPEED) # prepare map destination = carla.Transform() destination.location = world.get_random_location_from_navigation() #destination = get_random_destination(spawn_points) global_plan_map = replan(agent, destination, copy.deepcopy(origin_map)) # start to plan plan_thread = threading.Thread(target = make_plan, args=()) visualization_thread = threading.Thread(target = show_traj, args=()) while True: if (global_img is not None) and (global_nav is not None) and (global_pcd is not None): plan_thread.start() break else: time.sleep(0.001) # wait for the first plan result while not start_control: time.sleep(0.001) #visualization_thread.start() # start to control print('Start to control') ctrller = CapacController(world, vehicle, 30) while True: # change destination if close2dest(vehicle, destination): #destination = get_random_destination(spawn_points) print('get destination !', time.time()) destination = carla.Transform() destination.location = world.get_random_location_from_navigation() global_plan_map = replan(agent, destination, copy.deepcopy(origin_map)) v = global_vehicle.get_velocity() a = global_vehicle.get_acceleration() global_vel = np.sqrt(v.x**2+v.y**2+v.z**2) global_a = np.sqrt(a.x**2+a.y**2+a.z**2) #steer_angle = global_vehicle.get_control().steer*max_steer_angle #w = global_vel*np.tan(steer_angle)/2.405 control_time = time.time() dt = control_time - global_trajectory['time'] index = int((dt/args.max_t)//args.dt) if index > 0.99/args.dt: continue control = ctrller.run_step(global_trajectory, index, state0) vehicle.apply_control(control) curve = None#show_traj(True) visualize(global_view_img, draw_cost_map, global_nav, curve) #time.sleep(1/60.) cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): global global_nav, global_vel, start_control, global_plan_map, global_vehicle, global_transform, max_steer_angle, global_a, state0, global_collision client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_vehicle = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() max_steer_angle = np.deg2rad(physics_control.wheels[0].max_steer_angle) spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) agent = BasicAgent(vehicle, target_speed=MAX_SPEED) # prepare map destination = carla.Transform() destination.location = world.get_random_location_from_navigation() global_plan_map, destination = replan(agent, destination, copy.deepcopy(origin_map), spawn_points) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=6.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'collision': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': collision_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() time.sleep(0.5) # start to plan plan_thread = threading.Thread(target=make_plan, args=()) # visualization_thread = threading.Thread(target = show_traj, args=()) while True: if (global_img is not None) and (global_nav is not None): plan_thread.start() break else: time.sleep(0.001) # wait for the first plan result while not start_control: time.sleep(0.001) # visualization_thread.start() # start to control print('Start to control') ctrller = CapacController(world, vehicle, 30) while True: # change destination if close2dest(vehicle, destination): #destination = get_random_destination(spawn_points) print('get destination !', time.time()) destination = carla.Transform() destination.location = world.get_random_location_from_navigation() # global_plan_map = replan(agent, destination, copy.deepcopy(origin_map)) global_plan_map, destination = replan(agent, destination, copy.deepcopy(origin_map), spawn_points) if global_collision: global_collision = False start_point = random.choice(spawn_points) vehicle.set_transform(start_point) global_plan_map, destination = replan(agent, destination, copy.deepcopy(origin_map), spawn_points) start_waypoint = agent._map.get_waypoint( agent._vehicle.get_location()) end_waypoint = agent._map.get_waypoint(destination.location) route_trace = agent._trace_route(start_waypoint, end_waypoint) start_point.rotation = route_trace[0][0].transform.rotation vehicle.set_transform(start_point) v = global_vehicle.get_velocity() a = global_vehicle.get_acceleration() global_vel = np.sqrt(v.x**2 + v.y**2 + v.z**2) global_a = np.sqrt(a.x**2 + a.y**2 + a.z**2) #steer_angle = global_vehicle.get_control().steer*max_steer_angle #w = global_vel*np.tan(steer_angle)/2.405 control_time = time.time() dt = control_time - global_trajectory['time'] index = int((dt / args.max_t) // args.dt) + 2 if index > 0.99 / args.dt: continue """ transform = vehicle.get_transform() dx, dy, dyaw = get_transform(transform, global_transform) dyaw = -dyaw _x = global_trajectory['x'][index] - dx _y = global_trajectory['y'][index] - dy x = _x*np.cos(dyaw) + _y*np.sin(dyaw) y = _y*np.cos(dyaw) - _x*np.sin(dyaw) _vx = global_trajectory['vx'][index] _vy = global_trajectory['vy'][index] vx = _vx*np.cos(dyaw) + _vy*np.sin(dyaw) vy = _vy*np.cos(dyaw) - _vx*np.sin(dyaw) _ax = global_trajectory['ax'][index] _ay = global_trajectory['ay'][index] ax = _ax*np.cos(dyaw) + _ay*np.sin(dyaw) ay = _ay*np.cos(dyaw) - _ax*np.sin(dyaw) """ #control = get_control(x, y, vx, vy, ax, ay) control = ctrller.run_step(global_trajectory, index, state0) vehicle.apply_control(control) # x = global_trajectory['x'] # y = global_trajectory['y'] # plt.cla() # plt.plot(x, y) # plt.xlim(-1, 29) # plt.ylim(-15, 15) # plt.show() """ dyaw = np.deg2rad(global_transform.rotation.yaw) x = global_trajectory['x'][index]*np.cos(dyaw) + global_trajectory['y'][index]*np.sin(dyaw) y = global_trajectory['y'][index]*np.cos(dyaw) - global_trajectory['x'][index]*np.sin(dyaw) """ #localtion = carla.Location(x = global_transform.location.x+x, y=global_transform.location.y+y, z=2.0) #world.debug.draw_point(localtion, size=0.3, color=carla.Color(255,0,0), life_time=10.0) #print(global_vel*np.tan(control.steer)/w) curve = None #show_traj(True) visualize(global_view_img, global_nav, curve) #time.sleep(1/60.) cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): global global_nav, global_vel client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters( cloudiness=random.randint(0,10), precipitation=0, sun_altitude_angle=random.randint(50,90) ) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) sensor_dict = { 'camera':{ 'transform':carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback':image_callback, }, 'lidar':{ 'transform':carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback':lidar_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() time.sleep(0.3) spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) agent = BasicAgent(vehicle, target_speed=MAX_SPEED) destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, copy.deepcopy(origin_map)) inverse_perspective_mapping = InversePerspectiveMapping(param, sensor_master) time.sleep(0.3) while True: if close2dest(vehicle, destination): destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, copy.deepcopy(origin_map)) v = vehicle.get_velocity() global_vel = np.sqrt(v.x**2+v.y**2+v.z**2) global_nav = get_nav(vehicle, plan_map) result = get_cGAN_result() img = copy.deepcopy(global_img) mask = np.where(result > 200) img[mask[0],mask[1]] = (255, 0, 0, 255) ipm_image = inverse_perspective_mapping.getIPM(result) cost_map = get_cost_map(ipm_image, global_pcd) #plan_cost_map = draw_traj(cost_map) plan_cost_map, trajectory = get_traj(cost_map, 0.7, show=True) control = get_control(trajectory) vehicle.apply_control(control) visualize(img, plan_cost_map, global_nav) cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_dict['vehicle'] = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() global_dict['max_steer_angle'] = np.deg2rad( physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=9.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, 'collision': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': collision_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() env = CARLAEnv(world, vehicle, global_dict, args, trajectory_model) # start to control print('Start to control') state = env.reset() while True: state = env.reset() sm.close_all() vehicle.destroy()
client = carla.Client(args.host, args.port) client.set_timeout(2.0) display = pygame.display.set_mode((args.width, args.height), pygame.HWSURFACE | pygame.DOUBLEBUF) hud = HUD(args.width, args.height) _world = client.load_world('Town01') world_map = _world.get_map() world = World(_world, hud) weather = carla.WeatherParameters(cloudiness=random.randint(0, 80), precipitation=0, sun_altitude_angle=random.randint( 40, 90)) set_weather(_world, weather) controller = DualControl(world) vehicle = world.player agent = BasicAgent(vehicle, target_speed=31) spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, copy.deepcopy(origin_map)) #plan_map.resize((1000, 1000)).show() clock = pygame.time.Clock()
def main(): global global_nav, global_control, global_pos, global_vel client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) #world = client.get_world() world = client.load_world('Town02') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 70, 90)) #world.set_weather(carla.WeatherParameters.ClearNoon) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.bmw.grandtourer') # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=1.5, y=0.0, z=2.0)), 'callback': image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=1.5, y=0.0, z=2.0)), 'callback': lidar_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() time.sleep(0.3) spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) agent = BasicAgent(vehicle, target_speed=MAX_SPEED) destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, copy.deepcopy(origin_map)) FPS = [str(time.time())] for cnt in range(args.num): if close2dest(vehicle, destination): destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, copy.deepcopy(origin_map)) if vehicle.is_at_traffic_light(): traffic_light = vehicle.get_traffic_light() if traffic_light.get_state() == carla.TrafficLightState.Red: traffic_light.set_state(carla.TrafficLightState.Green) #control = agent.run_step() #control.manual_gear_shift = False #control = None #vehicle.apply_control(control) #nav = get_nav(vehicle, plan_map) global_pos = vehicle.get_transform() global_vel = vehicle.get_velocity() start_waypoint = agent._map.get_waypoint(agent._vehicle.get_location()) end_waypoint = agent._map.get_waypoint(destination.location) route_trace = agent._trace_route(start_waypoint, end_waypoint) x = [] y = [] traj_pose_list = [] for i in range(len(route_trace) - 1): waypoint1 = route_trace[i][0].transform.location waypoint2 = route_trace[i + 1][0].transform.location dist = waypoint1.distance(waypoint2) x.append(waypoint1.x) y.append(waypoint1.y) traj_pose_list.append((0, route_trace[i][0].transform)) plt.plot([global_pos.location.x, x[0]], [global_pos.location.y, y[0]], color='r') plt.plot(x[:10], y[:10], color='b') plt.show() empty_image = collect_perspective.getPM(traj_pose_list, global_pos, global_img) break #cv2.imshow('Nav', nav) #cv2.imshow('Vision', global_img) #cv2.waitKey(10) index = str(time.time()) if cnt % 100 == 0: FPS.append(index) print(round(1 / ((float(FPS[-1]) - float(FPS[-2])) / 100), 1)) cv2.destroyAllWindows() vehicle.destroy() sm.close_all()
def main(): client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_dict['vehicle'] = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() global_dict['max_steer_angle'] = np.deg2rad( physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=6.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() #spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) agent = BasicAgent(vehicle, target_speed=args.max_speed) # prepare map destination = carla.Transform() destination.location = world.get_random_location_from_navigation() #destination = get_random_destination(spawn_points) global_dict['plan_map'] = replan(agent, destination, copy.deepcopy(origin_map)) # start to plan plan_thread = threading.Thread(target=make_plan, args=()) #visualization_thread = threading.Thread(target = show_traj, args=()) while True: if (global_dict['img'] is not None) and (global_dict['nav'] is not None) and (global_dict['pcd'] is not None): plan_thread.start() break else: time.sleep(0.001) # wait for the first plan result while not global_dict['start_control']: time.sleep(0.001) #visualization_thread.start() # start to control print('Start to control') ctrller = CapacController(world, vehicle, 30) env = CARLAEnv(world, vehicle) env.reset() #env.step() while True: # change destination if close2dest(vehicle, destination): #destination = get_random_destination(spawn_points) print('get destination !', time.time()) destination = carla.Transform() destination.location = world.get_random_location_from_navigation() global_dict['plan_map'] = replan(agent, destination, copy.deepcopy(origin_map)) v = global_dict['vehicle'].get_velocity() a = global_dict['vehicle'].get_acceleration() global_dict['vel'] = np.sqrt(v.x**2 + v.y**2 + v.z**2) global_dict['a'] = np.sqrt(a.x**2 + a.y**2 + a.z**2) #steer_angle = global_dict['vehicle'].get_control().steer*global_dict['max_steer_angle'] #w = global_vel*np.tan(steer_angle)/2.405 control_time = time.time() dt = control_time - global_dict['trajectory']['time'] index = int((dt / args.max_t) // args.dt) + 5 if index > 0.99 / args.dt: continue """ transform = vehicle.get_transform() dx, dy, dyaw = get_diff_tf(transform, global_dict['transform']) dyaw = -dyaw _x = global_dict['trajectory']['x'][index] - dx _y = global_dict['trajectory']['y'][index] - dy x = _x*np.cos(dyaw) + _y*np.sin(dyaw) y = _y*np.cos(dyaw) - _x*np.sin(dyaw) _vx = global_dict['trajectory']['vx'][index] _vy = global_dict['trajectory']['vy'][index] vx = _vx*np.cos(dyaw) + _vy*np.sin(dyaw) vy = _vy*np.cos(dyaw) - _vx*np.sin(dyaw) _ax = global_dict['trajectory']['ax'][index] _ay = global_dict['trajectory']['ay'][index] ax = _ax*np.cos(dyaw) + _ay*np.sin(dyaw) ay = _ay*np.cos(dyaw) - _ax*np.sin(dyaw) """ control = ctrller.run_step(global_dict['trajectory'], index, global_dict['state0']) env.step(control) #vehicle.apply_control(control) """ dyaw = np.deg2rad(global_dict['transform'].rotation.yaw) x = global_dict['trajectory']['x'][index]*np.cos(dyaw) + global_dict['trajectory']['y'][index]*np.sin(dyaw) y = global_dict['trajectory']['y'][index]*np.cos(dyaw) - global_dict['trajectory']['x'][index]*np.sin(dyaw) """ curve = None #show_traj(True) #visualize(global_dict['view_img'], global_dict['draw_cost_map'], global_dict['nav'], args, curve) visualize(global_dict, global_dict['draw_cost_map'], args, curve) cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_dict['vehicle'] = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() global_dict['max_steer_angle'] = np.deg2rad( physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=9.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, 'collision': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': collision_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() ctrller = CapacController(world, vehicle, 30) agent = BasicAgent(vehicle, target_speed=30) world_map = world.get_map() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) spawn_points = world_map.get_spawn_points() route_trace = None start_point = random.choice(spawn_points) destination = random.choice(spawn_points) vehicle.set_transform(start_point) global_dict['plan_map'] = replan(agent, destination, copy.deepcopy(origin_map)) global_dict['collision'] = False start_waypoint = agent._map.get_waypoint(agent._vehicle.get_location()) end_waypoint = agent._map.get_waypoint(destination.location) route_trace = agent._trace_route(start_waypoint, end_waypoint) start_point.rotation = route_trace[0][0].transform.rotation vehicle.set_transform(start_point) plan_thread = threading.Thread(target=make_plan, args=()) while True: if (global_dict['img'] is not None) and (global_dict['nav'] is not None): plan_thread.start() break else: time.sleep(0.001) while global_dict['trajectory'] is None: time.sleep(0.001) while True: done = False trajectory = global_dict['trajectory'] for i in range(20): if global_dict['collision']: done = True print('Done !') break if close2dest(vehicle, destination): done = True print('Success !') break control_time = time.time() dt = control_time - trajectory['time'] index = int((dt / args.max_t) // args.dt) + 5 if index > 0.99 // args.dt - 10: continue control = ctrller.run_step(trajectory, index, global_dict['state0']) vehicle.apply_control(control) cv2.imshow('Visualization', global_dict['view_img']) cv2.imshow('Costmap', global_dict['draw_cost_map']) cv2.waitKey(1) if done or global_dict['collision']: start_point = random.choice(spawn_points) destination = random.choice(spawn_points) vehicle.set_transform(start_point) global_dict['plan_map'] = replan(agent, destination, copy.deepcopy(origin_map)) global_dict['collision'] = False start_waypoint = agent._map.get_waypoint( agent._vehicle.get_location()) end_waypoint = agent._map.get_waypoint(destination.location) route_trace = agent._trace_route(start_waypoint, end_waypoint) start_point.rotation = route_trace[0][0].transform.rotation vehicle.set_transform(start_point) done = False sm.close_all() vehicle.destroy()
def main(): global global_nav, global_vel, start_control, global_plan_map, global_vehicle, global_cost_map, global_transform, max_steer_angle, global_a, draw_cost_map client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_vehicle = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() max_steer_angle = np.deg2rad(physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() #spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) agent = BasicAgent(vehicle, target_speed=MAX_SPEED) # prepare map destination = carla.Transform() destination.location = world.get_random_location_from_navigation() #destination = get_random_destination(spawn_points) global_plan_map = replan(agent, destination, copy.deepcopy(origin_map)) # start to plan plan_thread = threading.Thread(target=make_plan, args=()) while True: if (global_img is not None) and (global_nav is not None) and (global_pcd is not None): plan_thread.start() break else: time.sleep(0.001) # wait for the first plan result while not start_control: time.sleep(0.001) # start to control print('Start to control') avg_dt = 1.0 distance = 0. last_location = vehicle.get_location() while True: # change destination if close2dest(vehicle, destination): #destination = get_random_destination(spawn_points) destination = carla.Transform() destination.location = world.get_random_location_from_navigation() global_plan_map = replan(agent, destination, copy.deepcopy(origin_map)) v = global_vehicle.get_velocity() a = global_vehicle.get_acceleration() global_vel = np.sqrt(v.x**2 + v.y**2 + v.z**2) global_a = np.sqrt(a.x**2 + a.y**2 + a.z**2) #steer_angle = global_vehicle.get_control().steer*max_steer_angle #w = global_vel*np.tan(steer_angle)/2.405 control_time = time.time() dt = control_time - global_trajectory['time'] avg_dt = 0.99 * avg_dt + 0.01 * dt #print(round(avg_dt, 3)) index = int((dt / args.max_t) // args.dt) if index > 0.9 / args.dt: continue location = vehicle.get_location() distance += location.distance(last_location) last_location = location #print(round(distance, 1)) transform = vehicle.get_transform() dx, dy, dyaw = get_transform(transform, global_transform) dyaw = -dyaw _x = global_trajectory['x'][index] - dx _y = global_trajectory['y'][index] - dy x = _x * np.cos(dyaw) + _y * np.sin(dyaw) y = _y * np.cos(dyaw) - _x * np.sin(dyaw) _vx = global_trajectory['vx'][index] _vy = global_trajectory['vy'][index] vx = _vx * np.cos(dyaw) + _vy * np.sin(dyaw) vy = _vy * np.cos(dyaw) - _vx * np.sin(dyaw) _ax = global_trajectory['ax'][index] _ay = global_trajectory['ay'][index] ax = _ax * np.cos(dyaw) + _ay * np.sin(dyaw) ay = _ay * np.cos(dyaw) - _ax * np.sin(dyaw) control = get_new_control(x, y, vx, vy, ax, ay) vehicle.apply_control(control) #print(global_vel*np.tan(control.steer)/w) #visualize(global_img, draw_cost_map, global_nav) visualize(global_img) #time.sleep(1/60.) cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.get_world() weather = carla.WeatherParameters(cloudiness=30.0, precipitation=30.0, sun_altitude_angle=50.0) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.bmw.grandtourer') # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=1.5, y=0.0, z=2.0)), 'callback': image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=1.5, y=0.0, z=2.0)), 'callback': lidar_callback, }, 'gnss': { 'transform': carla.Transform(carla.Location(x=0.0, y=0.0, z=0.0)), 'callback': gnss_callback, }, 'imu': { 'transform': carla.Transform(carla.Location(x=0.0, y=0.0, z=0.0)), 'callback': imu_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() time.sleep(0.3) vehicle.set_autopilot(True) try: while True: if vehicle.is_at_traffic_light(): traffic_light = vehicle.get_traffic_light() if traffic_light.get_state() == carla.TrafficLightState.Red: traffic_light.set_state(carla.TrafficLightState.Green) cv2.imshow('Raw image', global_img) """ vehicle.apply_control(carla.VehicleControl( throttle=0.3, steer = 0.0, reverse=False, manual_gear_shift=True, gear=1, brake=0.0, hand_brake=False )) """ #vehicle.get_angular_velocity().z #vehicle.get_velocity().x cv2.waitKey(16) #break cv2.destroyAllWindows() finally: sm.close_all() vehicle.destroy()
def main(): client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_dict['vehicle'] = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() global_dict['max_steer_angle'] = np.deg2rad( physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=9.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, 'collision': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': collision_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() model = TD3(buffer_size=3e3, noise_decay_steps=5e3, batch_size=10) env = CARLAEnv(world, vehicle, global_dict, args, trajectory_model) state = env.reset() plan_thread = threading.Thread(target=make_plan, args=()) while True: if (global_dict['img'] is not None) and (global_dict['nav'] is not None) and (global_dict['pcd'] is not None): plan_thread.start() break else: time.sleep(0.001) # start to control print('Start to control') episode_timesteps = 0 episode_reward = 0 max_steps = 1e9 total_steps = 0 max_episode_steps = 1000 learning_starts = 50 episode_num = 0 while total_steps < max_steps: total_steps += 1 episode_timesteps = 0 episode_reward = 0 state_queue = [] state = env.reset() for _ in range(10): state_queue.append(state) state = torch.stack(state_queue) plan_time = global_dict['plan_time'] for step in range(max_episode_steps): global_dict['state0'] = cu.getActorState('odom', plan_time, global_dict['vehicle']) global_dict['state0'].x = global_dict['transform'].location.x global_dict['state0'].y = global_dict['transform'].location.y global_dict['state0'].z = global_dict['transform'].location.z global_dict['state0'].theta = np.deg2rad( global_dict['transform'].rotation.yaw) action = model.policy_net.get_action(state) print('org:', action) action = model.noise.get_action(action, total_steps) print('after', action) next_state, reward, done, _ = env.step(action, plan_time) plan_time = global_dict['plan_time'] model.replay_buffer.push(copy.deepcopy(state), action, reward, copy.deepcopy(next_state), done) state = next_state episode_reward += reward total_steps += 1 episode_timesteps += 1 if done or episode_timesteps == max_episode_steps: if len(model.replay_buffer) > learning_starts: for i in range(episode_timesteps): model.train_step(i) episode_num += 1 if episode_num > 0 and episode_num % model.save_eps_num == 0: model.save(directory=model.load_dir, filename=str(episode_num)) model.writer.add_scalar('episode_reward', episode_reward, episode_num) break sm.close_all() vehicle.destroy()
def main(): client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() # world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.yamaha.yzf') #vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.*.*') global_dict['vehicle'] = vehicle # Enables or disables the simulation of physics on this actor. # vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() global_dict['max_steer_angle'] = np.deg2rad( physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=9.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, 'collision': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': collision_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() env = CARLAEnv(world, vehicle, global_dict, args, trajectory_model) plan_thread = threading.Thread(target=make_plan, args=()) while True: if (global_dict['img'] is not None) and (global_dict['nav'] is not None) and ( global_dict['trans_costmap'] is not None): plan_thread.start() break else: time.sleep(0.001) # start to control print('Start to control') state = env.reset() episode_timesteps = 0 episode_reward = 0 max_steps = 1e9 total_steps = 0 max_episode_steps = 1000 learning_starts = 10 #2000 episode_num = 0 last_episode_reward = 0 # log total_driving_metre = 0 while total_steps < max_steps: print("total_episode:", episode_num) episode_num += 1 total_steps += 1 episode_timesteps = 0 episode_reward = 0 total_driving_metre = 0 #state_queue = [] state = env.reset() #for _ in range(10): state_queue.append(state) #state = torch.stack(state_queue) plan_time = time.time() global_dict['state0'] = cu.getActorState('odom', plan_time, global_dict['vehicle']) global_dict['state0'].x = global_dict['transform'].location.x global_dict['state0'].y = global_dict['transform'].location.y global_dict['state0'].z = global_dict['transform'].location.z global_dict['state0'].theta = np.deg2rad( global_dict['transform'].rotation.yaw) add_noise = True if random.random() < 0.5 else False for step in range(max_episode_steps): action = model.policy_net.get_action(state) #print('org:', action) if add_noise: if last_episode_reward < 50: action = model.noise.get_action(action, 66) else: action = model.noise.get_action(action, total_steps // 2) # print('action:', action) x_last = global_dict['transform'].location.x y_last = global_dict['transform'].location.y next_state, reward, done, ts = env.step(action, plan_time) plan_time = ts global_dict['state0'] = cu.getActorState('odom', plan_time, global_dict['vehicle']) global_dict['state0'].x = global_dict['transform'].location.x global_dict['state0'].y = global_dict['transform'].location.y global_dict['state0'].z = global_dict['transform'].location.z global_dict['state0'].theta = np.deg2rad( global_dict['transform'].rotation.yaw) x_now = global_dict['transform'].location.x y_now = global_dict['transform'].location.y driving_metre_in_step = np.sqrt((x_now - x_last)**2 + (y_now - y_last)**2) total_driving_metre += driving_metre_in_step model.replay_buffer.push(state, action, reward, next_state, done) # print("next state :", type(next_state)) Tensor # print("next state shape:", next_state.shape) # print("state shape:", state.shape) next state shape: torch.Size([1, 200, 400]) if len(model.replay_buffer) > max(learning_starts, model.batch_size): print("Start Train") model.train_step(total_steps, noise_std=0.2, noise_clip=0.5) #noise_std = 0.2 state = next_state episode_reward += reward total_steps += 1 episode_timesteps += 1 if done or episode_timesteps == max_episode_steps: logger.add_scalar('episode_reward', episode_reward, episode_num) logger.add_scalar('total_driving_metre', total_driving_metre, episode_num) #if len(model.replay_buffer) > max(learning_starts, model.batch_size): # for i in range(episode_timesteps): # model.train_step(total_steps, noise_std = 0.1, noise_clip=0.25) # print(len(model.replay_buffer)) if episode_reward > 50: print('Success') else: print('Fail') last_episode_reward = episode_reward if episode_num % 50 == 0: model.save(directory=ckpt_path, filename=str(episode_num)) break sm.close_all() vehicle.destroy()
def main(): global global_transform, max_steer_angle client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) settings = world.get_settings() settings.synchronous_mode = True settings.fixed_delta_seconds = 0.01 world.apply_settings(settings) blueprint = world.get_blueprint_library() # world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') global_dict['vehicle'] = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() env = CARLAEnv(world, vehicle, global_dict, args) # state = env.reset() max_steer_angle = np.deg2rad(physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=6.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'collision': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': collision_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() time.sleep(0.5) while True: if (global_dict['img'] is not None) and (global_dict['nav'] is not None) and (global_dict['img_nav'] is not None): break else: world.tick() # time.sleep(0.01) print('Start to control') episode_timesteps = 0 episode_reward = 0 max_steps = 1e9 total_steps = 0 max_episode_steps = 1000 learning_starts = 2000 #2000 episode_num = 0 while total_steps < max_steps: print("total_episode:", episode_num) episode_num += 1 total_steps += 1 episode_timesteps = 0 episode_reward = 0 total_driving_metre = 0 state = env.reset() plan_time = time.time() global_dict['state0'] = cu.getActorState('odom', global_plan_time, global_dict['vehicle']) global_dict['state0'].x = global_transform.location.x global_dict['state0'].y = global_transform.location.y global_dict['state0'].z = global_transform.location.z global_dict['state0'].theta = np.deg2rad(global_transform.rotation.yaw) add_noise = True if random.random() < 0.5 else False for step in range(max_episode_steps): state_tensor = state.unsqueeze(0).to(device) action = model.policy_net(state_tensor) action = action.detach().cpu().numpy()[0] if add_noise: action = model.noise.get_action(action) x_last = global_transform.location.x y_last = global_transform.location.y next_state, reward, done, ts = env.step(action) plan_time = ts global_dict['state0'] = cu.getActorState('odom', global_plan_time, global_dict['vehicle']) global_dict['state0'].x = global_transform.location.x global_dict['state0'].y = global_transform.location.y global_dict['state0'].z = global_transform.location.z global_dict['state0'].theta = np.deg2rad( global_transform.rotation.yaw) x_now = global_transform.location.x y_now = global_transform.location.y driving_metre_in_step = np.sqrt((x_now - x_last)**2 + (y_now - y_last)**2) total_driving_metre += driving_metre_in_step model.replay_buffer.push(state, action, reward, next_state, done) if len(model.replay_buffer) > max(learning_starts, model.batch_size): print("Start Train") # time_s = time.time() model.train_step( total_steps, noise_std=0.2, noise_clip=0.3) #noise_std = 0.2 noise_clip = 0.5 # time_e = time.time() # print('time:', time_e - time_s) state = next_state episode_reward += reward total_steps += 1 episode_timesteps += 1 if done or episode_timesteps == max_episode_steps: logger.add_scalar('episode_reward', episode_reward, episode_num) logger.add_scalar('total_driving_metre', total_driving_metre, episode_num) #if len(model.replay_buffer) > max(learning_starts, model.batch_size): # for i in range(episode_timesteps): # model.train_step(total_steps, noise_std = 0.1, noise_clip=0.25) # print(len(model.replay_buffer)) if episode_reward > 50: print('Success') else: print('Fail') # last_episode_reward = episode_reward if episode_num % 5 == 0: model.save(directory=ckpt_path, filename=str(episode_num)) break cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): global global_transform, max_steer_angle client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) settings = world.get_settings() settings.synchronous_mode = True settings.fixed_delta_seconds = 0.05 world.apply_settings(settings) blueprint = world.get_blueprint_library() # world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') global_dict['vehicle'] = vehicle # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() env = CARLAEnv(world, vehicle, global_dict, args, generator) # state = env.reset() max_steer_angle = np.deg2rad(physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { #'transform':carla.Transform(carla.Location(x=0.0, y=4.0, z=4.0), carla.Rotation(pitch=-30, yaw=-60)), 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=6.0), carla.Rotation(pitch=-45)), #'transform':carla.Transform(carla.Location(x=0.0, y=0.0, z=6.0), carla.Rotation(pitch=-90)), 'callback': view_image_callback, }, 'collision': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': collision_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() time.sleep(0.5) while True: if (global_dict['img'] is not None) and (global_dict['nav'] is not None) and (global_dict['img_nav'] is not None): break else: world.tick() # time.sleep(0.01) # wait for the first plan result # while not start_control: # time.sleep(0.001) print('Start to control') # ctrller = CapacController(world, vehicle, 30) episode_timesteps = 0 episode_reward = 0 max_steps = 1e9 total_steps = 0 max_episode_steps = 1000 learning_starts = 1000 #2000 episode_num = 0 start_train = False while total_steps < max_steps: print("total_episode:", episode_num) episode_num += 1 total_steps += 1 episode_timesteps = 0 episode_reward = 0 total_driving_metre = 0 state = env.reset() plan_time = time.time() global_dict['state0'] = cu.getActorState('odom', global_plan_time, global_dict['vehicle']) global_dict['state0'].x = global_transform.location.x global_dict['state0'].y = global_transform.location.y global_dict['state0'].z = global_transform.location.z global_dict['state0'].theta = np.deg2rad(global_transform.rotation.yaw) add_noise = True if random.random() < 0.0 else False print("add noise:", add_noise) for step in range(max_episode_steps): # t = torch.arange(0, 0.99, args.dt).unsqueeze(1).to(device) # t.requires_grad = True t = np.arange(0, 0.99, args.dt) points_num = len(t) # v = global_dict['v0'] if global_dict['v0'] > 4 else 4 # v_0 = torch.FloatTensor([v/args.max_speed]).unsqueeze(1) # v_0 = v_0.to(device) if state['v0'] < 4: state['v0'] == 4 condition = torch.FloatTensor([state['v0'] / args.max_speed] * points_num).view(-1, 1) condition = condition.to(device) input_img = state['img_nav'].unsqueeze(0).to(device) v_0 = torch.FloatTensor([state['v0'] / args.max_speed ]).unsqueeze(1).to(device) action = model.policy_net(input_img, v_0) action = action.detach().cpu().numpy()[0] if add_noise: action = model.noise.get_action(action) # print(action) x_last = global_transform.location.x y_last = global_transform.location.y next_state, reward, done, ts = env.step(action, plan_time, condition) plan_time = ts global_dict['state0'] = cu.getActorState('odom', global_plan_time, global_dict['vehicle']) global_dict['state0'].x = global_transform.location.x global_dict['state0'].y = global_transform.location.y global_dict['state0'].z = global_transform.location.z global_dict['state0'].theta = np.deg2rad( global_transform.rotation.yaw) x_now = global_transform.location.x y_now = global_transform.location.y driving_metre_in_step = np.sqrt((x_now - x_last)**2 + (y_now - y_last)**2) total_driving_metre += driving_metre_in_step model.replay_buffer.push(state, action, reward, next_state, done) # print(sys.getsizeof(model.replay_buffer.buffer)) # print(u'当前进程的内存使用:%.4f GB' % (psutil.Process(os.getpid()).memory_info().rss / 1024 / 1024 / 1024) ) if len(model.replay_buffer) > max(learning_starts, model.batch_size): start_train = True print("Start Train:") # time_s = time.time() model.train_step( total_steps, noise_std=0.5, noise_clip=2) #noise_std = 0.2 noise_clip = 0.5 # time_e = time.time() # print('time:', time_e - time_s) state = next_state episode_reward += reward total_steps += 1 episode_timesteps += 1 if done or episode_timesteps == max_episode_steps: logger.add_scalar('episode_reward', episode_reward, episode_num) logger.add_scalar('total_driving_metre', total_driving_metre, episode_num) #if len(model.replay_buffer) > max(learning_starts, model.batch_size): # for i in range(episode_timesteps): # model.train_step(total_steps, noise_std = 0.1, noise_clip=0.25) # print(len(model.replay_buffer)) print("episode_reward:%.2f" % (episode_reward)) # if episode_reward > 50: # print('Success') # else: # print('Fail') # last_episode_reward = episode_reward if episode_num % 20 == 0 and start_train == True: model.save(directory=ckpt_path, filename=str(episode_num)) break cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): global global_nav, global_control, global_pos, global_vel client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) #world = client.get_world() world = client.load_world('Town10') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) #world.set_weather(carla.WeatherParameters.ClearNoon) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') # Enables or disables the simulation of physics on this actor. vehicle.set_simulate_physics(True) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=3.5), carla.Rotation(pitch=-30)), 'callback': view_image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() time.sleep(0.3) spawn_points = world_map.get_spawn_points() waypoint_tuple_list = world_map.get_topology() origin_map = get_map(waypoint_tuple_list) agent = BasicAgent(vehicle, target_speed=MAX_SPEED) #port = 8000 #tm = client.get_trafficmanager(port) #vehicle.set_autopilot(True,port) #tm.ignore_lights_percentage(vehicle,100) destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, copy.deepcopy(origin_map)) joystick = JoyStick(verbose=False) joystick.start() for cnt in tqdm(range(args.num)): if close2dest(vehicle, destination): destination = get_random_destination(spawn_points) plan_map = replan(agent, destination, copy.deepcopy(origin_map)) """ if vehicle.is_at_traffic_light(): traffic_light = vehicle.get_traffic_light() if traffic_light.get_state() == carla.TrafficLightState.Red: traffic_light.set_state(carla.TrafficLightState.Green) """ #control = agent.run_step() control = joystick.get() velocity = vehicle.get_velocity() if np.sqrt(velocity.x**2 + velocity.y**2) > 8: control.throttle = 0 #control.manual_gear_shift = False global_control = control vehicle.apply_control(control) nav = get_nav(vehicle, plan_map) global_nav = nav global_pos = vehicle.get_transform() global_vel = vehicle.get_velocity() #cv2.imshow('Nav', nav) cv2.imshow('Vision', global_view_img) cv2.waitKey(10) index = str(time.time()) #save_data(index) cmd_file.close() pos_file.close() vel_file.close() cv2.destroyAllWindows() sm.close_all() vehicle.destroy()
def main(): global global_img, global_view_img, global_pcd client = carla.Client(config['host'], config['port']) client.set_timeout(config['timeout']) world = client.load_world('Town01') weather = carla.WeatherParameters(cloudiness=random.randint(0, 10), precipitation=0, sun_altitude_angle=random.randint( 50, 90)) set_weather(world, weather) blueprint = world.get_blueprint_library() world_map = world.get_map() vehicle = add_vehicle(world, blueprint, vehicle_type='vehicle.audi.a2') vehicle.set_simulate_physics(True) physics_control = vehicle.get_physics_control() #max_steer_angle = np.deg2rad(physics_control.wheels[0].max_steer_angle) sensor_dict = { 'camera': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': image_callback, }, 'camera:view': { 'transform': carla.Transform(carla.Location(x=-3.0, y=0.0, z=8.0), carla.Rotation(pitch=-45)), 'callback': view_image_callback, }, 'lidar': { 'transform': carla.Transform(carla.Location(x=0.5, y=0.0, z=2.5)), 'callback': lidar_callback, }, } sm = SensorManager(world, blueprint, vehicle, sensor_dict) sm.init_all() #agent = BasicAgent(vehicle, target_speed=MAX_SPEED) destination = carla.Transform() destination.location = world.get_random_location_from_navigation() env = CARLAEnv() env.reset() env.step() # start to control print('Start to control') vehicle.set_autopilot(True) while True: # change destination if close2dest(vehicle, destination): destination = carla.Transform() destination.location = world.get_random_location_from_navigation() #control = None #vehicle.apply_control(control) #localtion = carla.Location(x = global_transform.location.x+x, y=global_transform.location.y+y, z=2.0) #world.debug.draw_point(localtion, size=0.3, color=carla.Color(255,0,0), life_time=10.0) forward_vector = vehicle.get_transform().get_forward_vector() begin = vehicle.get_location() end = vehicle.get_location() end.x = end.x + forward_vector.x * 8 end.y = end.y + forward_vector.y * 8 end.z = end.z + forward_vector.z * 8 world.debug.draw_line(begin, end, thickness=0.2, color=carla.Color(r=255, g=0, b=0, a=128), life_time=0.1) cv2.imshow('Visualization', global_view_img) cv2.waitKey(5) #time.sleep(1/60.) cv2.destroyAllWindows() sm.close_all() vehicle.destroy()