UtilityFlowSim/simulation.py at main · xPEDRAMx/UtilityFlowSim · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
# simulation.py

import random
import copy
import numpy as np
from config import CONFIG
from environment import ROADS, OD_LANE_KEEPING_ROADS, SIDEWALKS, DRONE_STATIONS, BIKE_LANES
from agents import Agent
import matplotlib.pyplot as plt
from visualization import update_live_plot


SPAWN_DATA = {
    "VEHICLE_SPAWN_POINTS": {
        'I1_NBO': lambda: (random.uniform(ROADS['I1_NBO']['x_min'], ROADS['I1_NBO']['x_max']), ROADS['I1_NBO']['y_min'] + 5, np.pi/2),
        'I1_SBO': lambda: (random.uniform(ROADS['I1_SBO']['x_min'], ROADS['I1_SBO']['x_max']), ROADS['I1_SBO']['y_max'] - 5, -np.pi/2),
        'I1_EBO': lambda: (ROADS['I1_EBO']['x_min'] + 5, random.uniform(ROADS['I1_EBO']['y_min'], ROADS['I1_EBO']['y_max']), 0),
        'I1_WBO': lambda: (ROADS['I1_WBO']['x_max'] - 5, random.uniform(ROADS['I1_WBO']['y_min'], ROADS['I1_WBO']['y_max']), np.pi),
        'I2_NBO': lambda: (random.uniform(ROADS['I2_NBO']['x_min'], ROADS['I2_NBO']['x_max']), ROADS['I2_NBO']['y_min'] + 5, np.pi/2),
        'I2_SBO': lambda: (random.uniform(ROADS['I2_SBO']['x_min'], ROADS['I2_SBO']['x_max']), ROADS['I2_SBO']['y_max'] - 5, -np.pi/2),
        'I2_WBO': lambda: (ROADS['I2_WBO']['x_max'] - 5, random.uniform(ROADS['I2_WBO']['y_min'], ROADS['I2_WBO']['y_max']), np.pi),
    },
    "VEHICLE_DEST_POINTS": {
        'I1_NBD': lambda: (random.uniform(ROADS['I1_NBD']['x_min'], ROADS['I1_NBD']['x_max']), ROADS['I1_NBD']['y_max']),
        'I1_SBD': lambda: (random.uniform(ROADS['I1_SBD']['x_min'], ROADS['I1_SBD']['x_max']), ROADS['I1_SBD']['y_min']),
        'I1_EBD': lambda: (ROADS['I1_EBD']['x_max'], random.uniform(ROADS['I1_EBD']['y_min'], ROADS['I1_EBD']['y_max'])),
        'I1_WBD': lambda: (ROADS['I1_WBD']['x_min'], random.uniform(ROADS['I1_WBD']['y_min'], ROADS['I1_WBD']['y_max'])),
        'I2_NBD': lambda: (random.uniform(ROADS['I2_NBD']['x_min'], ROADS['I2_NBD']['x_max']), ROADS['I2_NBD']['y_max']),
        'I2_SBD': lambda: (random.uniform(ROADS['I2_SBD']['x_min'], ROADS['I2_SBD']['x_max']), ROADS['I2_SBD']['y_min']),
        'I2_EBD': lambda: (ROADS['I2_EBD']['x_max'], random.uniform(ROADS['I2_EBD']['y_min'], ROADS['I2_EBD']['y_max'])),
    },
    "VALID_OD_PAIRS": list(OD_LANE_KEEPING_ROADS.keys()),
    "SIDEWALK_KEYS": list(SIDEWALKS.keys()),
    "BIKE_LANE_KEYS": list(BIKE_LANES.keys()), # ADDED
    "STATION_KEYS": list(DRONE_STATIONS.keys())
}

def _create_single_agent(agent_type, agent_id, current_agents, config, spawn_data, creation_time):
    for _ in range(50):
        if agent_type == 'vehicle':
            if not spawn_data["VALID_OD_PAIRS"]: return None
            origin_key, dest_key = random.choice(spawn_data["VALID_OD_PAIRS"])
            initial_pos_x, initial_pos_y, initial_heading = spawn_data["VEHICLE_SPAWN_POINTS"][origin_key]()
            initial_pos = np.array([initial_pos_x, initial_pos_y, 0.0])
            if any(np.linalg.norm(initial_pos[:2] - a.pos[:2]) < config["vehicle_length"] for a in current_agents):
                continue
            dest_pos_x, dest_pos_y = spawn_data["VEHICLE_DEST_POINTS"][dest_key]()
            destination_pos = [dest_pos_x, dest_pos_y, 0.0]
            v_min, v_max = config["vehicle_speed_range"]
            agent_params = config.copy()
            agent_params["desired_speed_agent"] = random.uniform(v_min, v_max)
            agent_params["origin_key"], agent_params["destination_key"] = origin_key, dest_key
            return Agent(id=agent_id, agent_type='vehicle', initial_pos=initial_pos, initial_heading=initial_heading, initial_speed=random.uniform(v_min, v_max * 0.8), destination_pos=destination_pos, params=agent_params, creation_time=creation_time)

        elif agent_type == 'bike':
            agent_params = config.copy()
            bike_mode = config.get("bike_behavior_mode", "road")
            is_setting_2 = config.get("environment_setting", 1) == 2

            # In Setting 2, "sidewalk" mode is re-purposed to be "bike_lane" mode.
            if bike_mode == "sidewalk" and is_setting_2:
                if len(spawn_data["BIKE_LANE_KEYS"]) < 2: return None
                origin_bl_key, dest_bl_key = random.sample(spawn_data["BIKE_LANE_KEYS"], 2)
                origin_bounds, dest_bounds = BIKE_LANES[origin_bl_key], BIKE_LANES[dest_bl_key]
                initial_pos = np.array([random.uniform(origin_bounds['x_min'], origin_bounds['x_max']), random.uniform(origin_bounds['y_min'], origin_bounds['y_max']), 0.0])
                if any(np.linalg.norm(initial_pos[:2] - a.pos[:2]) < config['bb_safety_distance'] for a in current_agents):
                    continue
                destination_pos = [random.uniform(dest_bounds['x_min'], dest_bounds['x_max']), random.uniform(dest_bounds['y_min'], dest_bounds['y_max']), 0.0]
                initial_heading = np.arctan2(destination_pos[1] - initial_pos[1], destination_pos[0] - initial_pos[0])
                agent_params["origin_key"], agent_params["destination_key"] = None, None

            elif bike_mode == "road":
                if not spawn_data["VALID_OD_PAIRS"]: return None
                origin_key, dest_key = random.choice(spawn_data["VALID_OD_PAIRS"])
                initial_pos_x, initial_pos_y, initial_heading = spawn_data["VEHICLE_SPAWN_POINTS"][origin_key]()
                initial_pos = np.array([initial_pos_x, initial_pos_y, 0.0])
                if any(np.linalg.norm(initial_pos[:2] - a.pos[:2]) < config["bike_length"] for a in current_agents):
                    continue
                dest_pos_x, dest_pos_y = spawn_data["VEHICLE_DEST_POINTS"][dest_key]()
                destination_pos = [dest_pos_x, dest_pos_y, 0.0]
                agent_params["origin_key"], agent_params["destination_key"] = origin_key, dest_key

            else: # Sidewalk mode in Setting 1 (or if bike lanes don't exist)
                if len(spawn_data["SIDEWALK_KEYS"]) < 2: return None
                origin_sw_key, dest_sw_key = random.sample(spawn_data["SIDEWALK_KEYS"], 2)
                origin_bounds, dest_bounds = SIDEWALKS[origin_sw_key], SIDEWALKS[dest_sw_key]
                initial_pos = np.array([random.uniform(origin_bounds['x_min'], origin_bounds['x_max']), random.uniform(origin_bounds['y_min'], origin_bounds['y_max']), 0.0])
                if any(np.linalg.norm(initial_pos[:2] - a.pos[:2]) < config['pb_safety_distance'] for a in current_agents):
                    continue
                destination_pos = [random.uniform(dest_bounds['x_min'], dest_bounds['x_max']), random.uniform(dest_bounds['y_min'], dest_bounds['y_max']), 0.0]
                initial_heading = np.arctan2(destination_pos[1] - initial_pos[1], destination_pos[0] - initial_pos[0])
                agent_params["origin_key"], agent_params["destination_key"] = None, None

            v_min, v_max = config["bike_speed_range"]
            agent_params["desired_speed_agent"] = random.uniform(v_min, v_max)
            return Agent(id=agent_id, agent_type='bike', initial_pos=initial_pos, initial_heading=initial_heading, initial_speed=random.uniform(v_min, v_max * 0.8), destination_pos=destination_pos, params=agent_params, creation_time=creation_time)

        elif agent_type == 'pedestrian':
            if len(spawn_data["SIDEWALK_KEYS"]) < 2: return None
            origin_sw_key, dest_sw_key = random.sample(spawn_data["SIDEWALK_KEYS"], 2)
            origin_bounds, dest_bounds = SIDEWALKS[origin_sw_key], SIDEWALKS[dest_sw_key]
            initial_pos = np.array([random.uniform(origin_bounds['x_min'], origin_bounds['x_max']), random.uniform(origin_bounds['y_min'], origin_bounds['y_max']), 0.0])
            if any(np.linalg.norm(initial_pos[:2] - a.pos[:2]) < config['pp_safety_distance'] for a in current_agents):
                continue
            destination_pos = [random.uniform(dest_bounds['x_min'], dest_bounds['x_max']), random.uniform(dest_bounds['y_min'], dest_bounds['y_max']), 0.0]
            initial_heading = np.arctan2(destination_pos[1] - initial_pos[1], destination_pos[0] - initial_pos[0])
            p_min, p_max = config["pedestrian_speed_range"]
            return Agent(id=agent_id, agent_type='pedestrian', initial_pos=initial_pos, initial_heading=initial_heading, initial_speed=random.uniform(p_min, p_max), destination_pos=destination_pos, params=config.copy(), creation_time=creation_time)

        elif agent_type == 'drone':
            if not spawn_data["SIDEWALK_KEYS"] or not spawn_data["STATION_KEYS"]: return None
            origin_sw_key = random.choice(spawn_data["SIDEWALK_KEYS"])
            origin_bounds = SIDEWALKS[origin_sw_key]
            initial_pos = np.array([random.uniform(origin_bounds['x_min'], origin_bounds['x_max']), random.uniform(origin_bounds['y_min'], origin_bounds['y_max']), 0.0])
            if any(np.linalg.norm(initial_pos[:2] - a.pos[:2]) < config['dd_safety_distance'] for a in current_agents):
                continue
            dest_key = random.choice(spawn_data["STATION_KEYS"])
            destination_pos = DRONE_STATIONS[dest_key]['pos']
            initial_heading = np.arctan2(destination_pos[1] - initial_pos[1], destination_pos[0] - initial_pos[0])
            d_min, d_max = config["drone_speed_range"]
            agent_params = config.copy()
            agent_params["origin_key"], agent_params["destination_key"] = origin_sw_key, dest_key
            return Agent(id=agent_id, agent_type='drone', initial_pos=initial_pos, initial_heading=initial_heading, initial_speed=random.uniform(d_min, d_max), destination_pos=destination_pos, params=agent_params, creation_time=creation_time)

    print(f"Warning: Could not place new {agent_type} without overlap after 50 attempts.")
    return None

def create_agents(config):
    agents = []
    agent_id_counter = 0
    initial_counts = config.get("initial_agent_counts", {})
    for agent_type, count in initial_counts.items():
        for _ in range(count):
            new_agent = _create_single_agent(agent_type, agent_id_counter, agents, config, SPAWN_DATA, creation_time=0.0)
            if new_agent:
                agents.append(new_agent)
                agent_id_counter += 1
    return agents

def run_simulation_loop(initial_agents, config):
    num_time_steps = int(config["total_simulation_time"] / config["dt"])
    all_agents_for_history = list(initial_agents)
    active_agents = list(initial_agents)
    agent_id_counter = len(initial_agents)
    time_since_last_spawn = 0.0
    dynamic_generation_enabled = config.get("enable_dynamic_generation", False)
    generation_interval = config.get("agent_generation_interval", 10.0)
    fig, ax = None, None
    if config.get("show_live_simulation", False):
        fig, ax = plt.subplots(figsize=(14, 14)); plt.ion(); plt.show(block=False)
    print(f"\nStarting simulation for {config['total_simulation_time']}s ({num_time_steps} steps).")
    progress_stride = max(1, num_time_steps // 10)
    for t_step in range(num_time_steps):
        current_time = t_step * config["dt"]
        # Keep active set clean: completed agents are removed from simulation updates.
        active_agents = [agent for agent in active_agents if not agent.reached_destination]
        if dynamic_generation_enabled:
            time_since_last_spawn += config["dt"]
            if time_since_last_spawn >= generation_interval:
                time_since_last_spawn -= generation_interval
                generation_counts = config.get("dynamic_generation_counts", {})
                for agent_type, count in generation_counts.items():
                    for _ in range(count):
                        new_agent = _create_single_agent(agent_type, agent_id_counter, active_agents, config, SPAWN_DATA, creation_time=current_time)
                        if new_agent:
                            all_agents_for_history.append(new_agent)
                            active_agents.append(new_agent)
                            agent_id_counter += 1
        if (t_step + 1) % progress_stride == 0 or t_step == 0:
            print(f"Simulating step {t_step+1}/{num_time_steps} (Time: {current_time:.1f}s) - Total Agents: {len(all_agents_for_history)}, Active: {len(active_agents)}")
        if not active_agents and not dynamic_generation_enabled:
            print("All agents reached destination."); break
        # Synchronous decisions: all agents react to the same time-t snapshot.
        agent_state_snapshot = copy.deepcopy(active_agents)
        for agent_obj in active_agents:
            agent_obj.decide_and_move(agent_state_snapshot, current_time)
        active_agents = [agent for agent in active_agents if not agent.reached_destination]
        if fig and ax:
            update_live_plot(ax, active_agents, config, current_time); plt.pause(0.001)
    if fig and ax:
        plt.ioff(); plt.show()
    print("Simulation finished.")
    return all_agents_for_history