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
184
185
186
187
188
| """
BMW Symbiotic Drive 驾驶员意图推断模型
核心思路：
1. 融合方向盘、眼神、转向灯、身体姿态
2. 判断当前操作是"刻意"还是"无意识"
3. 刻意 → 抑制系统干预
4. 无意识 → 触发警告/干预
"""
import numpy as np
from dataclasses import dataclass
from typing import Optional, List
from enum import Enum
class DriverIntent(Enum):
DELIBERATE_LANE_CHANGE = "deliberate_lane_change"
UNCONSCIOUS_DEVIATION = "unconscious_deviation"
EMERGENCY_AVOIDANCE = "emergency_avoidance"
NORMAL_DRIVING = "normal_driving"
PARKING = "parking"
@dataclass
class IntentSignals:
"""多模态意图信号"""
steering_angle: float
steering_velocity: float
steering_jerk: float
gaze_zone: str
gaze_on_mirror: bool
gaze_on_blind_spot: bool
turn_signal_active: bool
turn_signal_direction: str
shoulder_check: bool
body_rotation: float
lane_departure: bool
vehicle_speed: float
time_to_lane_crossing: float
class DriverIntentInference:
"""
驾驶员意图推断引擎
BMW Symbiotic Drive 核心逻辑：
- 转向灯 + 肩部检查 + 眼神看盲区 → 刻意变道，抑制警告
- 无转向灯 + 无肩部检查 + 无眼神移动 → 无意识偏离，触发警告
- 急打方向 + 前方有障碍 → 紧急避险，增强辅助
"""
def __init__(self):
self.deliberate_threshold = 0.7
self.emergency_steering_vel = 300
self.lane_crossing_time_threshold = 2.0
def infer(self, signals: IntentSignals) -> dict:
"""推断驾驶员意图"""
intent_scores = {
DriverIntent.DELIBERATE_LANE_CHANGE: 0.0,
DriverIntent.UNCONSCIOUS_DEVIATION: 0.0,
DriverIntent.EMERGENCY_AVOIDANCE: 0.0,
DriverIntent.NORMAL_DRIVING: 0.0,
DriverIntent.PARKING: 0.0,
}
score = 0.0
if signals.turn_signal_active:
score += 0.35
if signals.shoulder_check:
score += 0.25
if signals.gaze_on_blind_spot:
score += 0.25
if signals.gaze_on_mirror:
score += 0.15
if abs(signals.steering_velocity) > 10 and abs(signals.steering_jerk) < 50:
score += 0.15
intent_scores[DriverIntent.DELIBERATE_LANE_CHANGE] = min(1.0, score)
score = 0.0
if signals.lane_departure:
score += 0.3
if not signals.turn_signal_active:
score += 0.25
if not signals.shoulder_check:
score += 0.2
if signals.gaze_zone == 'phone' or signals.gaze_zone == 'infotainment':
score += 0.25
if signals.time_to_lane_crossing < self.lane_crossing_time_threshold:
score += 0.2
intent_scores[DriverIntent.UNCONSCIOUS_DEVIATION] = min(1.0, score)
score = 0.0
if abs(signals.steering_velocity) > self.emergency_steering_vel:
score += 0.5
if abs(signals.steering_jerk) > 100:
score += 0.3
if signals.vehicle_speed > 60:
score += 0.2
intent_scores[DriverIntent.EMERGENCY_AVOIDANCE] = min(1.0, score)
score = 0.3
if not signals.lane_departure:
score += 0.3
if signals.gaze_zone == 'road':
score += 0.2
if abs(signals.steering_velocity) < 15:
score += 0.2
intent_scores[DriverIntent.NORMAL_DRIVING] = min(1.0, score)
if signals.vehicle_speed < 10:
intent_scores[DriverIntent.PARKING] = 0.8
for k in intent_scores:
intent_scores[k] *= 0.5
total = sum(intent_scores.values())
if total > 0:
for k in intent_scores:
intent_scores[k] /= total
best_intent = max(intent_scores, key=intent_scores.get)
confidence = intent_scores[best_intent]
if best_intent == DriverIntent.DELIBERATE_LANE_CHANGE and confidence > self.deliberate_threshold:
action = "SUPPRESS_WARNING"
elif best_intent == DriverIntent.UNCONSCIOUS_DEVIATION:
action = "TRIGGER_WARNING"
elif best_intent == DriverIntent.EMERGENCY_AVOIDANCE:
action = "ASSIST_EMERGENCY"
else:
action = "MONITOR"
return {
'intent': best_intent.value,
'confidence': confidence,
'all_scores': {k.value: v for k, v in intent_scores.items()},
'action': action,
}
if __name__ == "__main__":
engine = DriverIntentInference()
s1 = IntentSignals(
steering_angle=15, steering_velocity=20, steering_jerk=30,
gaze_zone='blind_spot', gaze_on_mirror=False, gaze_on_blind_spot=True,
turn_signal_active=True, turn_signal_direction='left',
shoulder_check=True, body_rotation=25,
lane_departure=False, vehicle_speed=80, time_to_lane_crossing=5
)
r1 = engine.infer(s1)
print(f"刻意变道: {r1['intent']}, conf={r1['confidence']:.2%}, action={r1['action']}")
s2 = IntentSignals(
steering_angle=5, steering_velocity=8, steering_jerk=15,
gaze_zone='phone', gaze_on_mirror=False, gaze_on_blind_spot=False,
turn_signal_active=False, turn_signal_direction='none',
shoulder_check=False, body_rotation=0,
lane_departure=True, vehicle_speed=70, time_to_lane_crossing=1.5
)
r2 = engine.infer(s2)
print(f"无意识偏离: {r2['intent']}, conf={r2['confidence']:.2%}, action={r2['action']}")
s3 = IntentSignals(
steering_angle=45, steering_velocity=350, steering_jerk=200,
gaze_zone='road', gaze_on_mirror=False, gaze_on_blind_spot=False,
turn_signal_active=False, turn_signal_direction='none',
shoulder_check=False, body_rotation=0,
lane_departure=False, vehicle_speed=90, time_to_lane_crossing=10
)
r3 = engine.infer(s3)
print(f"紧急避险: {r3['intent']}, conf={r3['confidence']:.2%}, action={r3['action']}")
|