/
toy_av_simulator.py
475 lines (389 loc) · 16.7 KB
/
toy_av_simulator.py
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
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
"""A toy simulator of a scenario of an AV approaching a crosswalk where some pedestrians are crossing."""
import pdb # Used for debugging
import numpy as np # Used for math
# Define the class
class ToyAVSimulator():
"""A toy simulator of a scenario of an AV approaching a crosswalk where some pedestrians are crossing.
The vehicle runs a modified version of the Intelligent Driver Model [1]_. The vehicle treats the closest
pedestrian in the road as a car to follow. If no pedestrians are in the road, it attempts to maintain the
desired speed. Noisy observations of the pedestrian are smoothed through an alpha-beta filter [2]_.
A collision results if any pedestrian's x-distance and y-distance to the ego vehicle are less than the respective
`min_dist_x` and `min_dist_y`.
The origin is centered in the middle of the east/west lane and the north/south crosswalk.
The positive x proceeds east down the lane, the positive y proceeds north across the crosswalk.
Parameters
----------
num_peds : int
The number of pedestrians crossing the street.
dt : float
The length (in seconds) of each timestep.
alpha : float
The alpha parameter in the tracker's alpha-beta filter [2]_.
beta : float
The beta parameter in the tracker's alpha-beta filter [2]_.
v_des : float
The desired velocity, in meters per second, for the ego vehicle to maintain
delta : float
The delta parameter in the IDM algorithm [1]_.
t_headway : float
The headway parameter in the IDM algorithm [1]_.
a_max : float
The maximum acceleration parameter in the IDM algorithm [1]_.
s_min : float
The minimum follow distance parameter in the IDM algorithm [1]_.
d_cmf : float
The maximum comfortable deceleration parameter in the IDM algorithm [1]_.
d_max : float
The maximum deceleration parameter in the IDM algorithm [1]_.
min_dist_x : float
The minimum x-distance between the ego vehicle and a pedestrian.
min_dist_y : float
The minimum y-distance between the ego vehicle and a pedestrian.
car_init_x : float
The initial x-position of the ego vehicle.
car_init_y : float
The initial y-position of the ego vehicle.
References
----------
.. [1] Treiber, Martin, Ansgar Hennecke, and Dirk Helbing.
"Congested traffic states in empirical observations and microscopic simulations."
Physical review E 62.2 (2000): 1805.
`<https://journals.aps.org/pre/abstract/10.1103/PhysRevE.62.1805>`_
.. [2] Rogers, Steven R. "Alpha-beta filter with correlated measurement noise."
IEEE Transactions on Aerospace and Electronic Systems 4 (1987): 592-594.
`<https://ieeexplore.ieee.org/abstract/document/4104388>`_
"""
# Accept parameters for defining the behavior of the system under test[SUT]
def __init__(self,
num_peds=1,
dt=0.1,
alpha=0.85,
beta=0.005,
v_des=11.17,
delta=4.0,
t_headway=1.5,
a_max=3.0,
s_min=4.0,
d_cmf=2.0,
d_max=9.0,
min_dist_x=2.5,
min_dist_y=1.4,
car_init_x=-35.0,
car_init_y=0.0,
):
# Constant hyper-params -- set by user
self.c_num_peds = num_peds
self.c_dt = dt
self.c_alpha = alpha
self.c_beta = beta
self.c_v_des = v_des
self.c_delta = delta
self.c_t_headway = t_headway
self.c_a_max = a_max
self.c_s_min = s_min
self.c_d_cmf = d_cmf
self.c_d_max = d_max
self.c_min_dist = np.array([min_dist_x, min_dist_y])
self.c_car_init_x = car_init_x
self.c_car_init_y = car_init_y
# self.blackbox_sim_state = blackbox_sim_state
# These are set by reset, not the user
self._car = np.zeros((4))
self._car_accel = np.zeros((2))
self._peds = np.zeros((self.c_num_peds, 4))
self._measurements = np.zeros((self.c_num_peds, 4))
self._car_obs = np.zeros((self.c_num_peds, 4))
self._env_obs = np.zeros((self.c_num_peds, 4))
self._done = False
self._reward = 0.0
self._info = []
self._step = 0
self._path_length = 0
# self._action = None
self._action = np.array([0] * (6 * self.c_num_peds))
self._first_step = True
self.directions = np.random.randint(2, size=self.c_num_peds) * 2 - 1
self.y = np.random.rand(self.c_num_peds) * 14 - 5
self.x = np.random.rand(self.c_num_peds) * 4 - 2
self._state = None
def run_simulation(self, actions, s_0, simulation_horizon):
"""Run a full simulation given the AST solver's actions and initial conditions.
Parameters
----------
actions : list[array_like]
A sequential list of actions taken by the AST Solver which deterministically control the simulation.
s_0 : array_like
An array specifying the initial conditions to set the simulator to.
simulation_horizon : int
The maximum number of steps a simulation rollout is allowed to run.
Returns
-------
terminal_index : int
The index of the action that resulted in a state in the goal set E. If no state is found
terminal_index should be returned as -1.
array_like
An array of relevant simulator info, which can then be used for analysis or diagnostics.
"""
# initialize the simulation
path_length = 0
self.reset(s_0)
self._info = []
simulation_horizon = np.minimum(simulation_horizon, len(actions))
# Take simulation steps unbtil horizon is reached
while path_length < simulation_horizon:
# get the action from the list
self._action = actions[path_length]
# Step the simulation forward in time
self.step_simulation(self._action)
# check if a crash has occurred. If so return the timestep, otherwise continue
if self.collision_detected():
return path_length, np.array(self._info)
path_length = path_length + 1
# horizon reached without crash, return -1
self._is_terminal = True
return -1, np.array(self._info)
def step_simulation(self, action):
"""
Handle anything that needs to take place at each step, such as a simulation update or write to file.
Parameters
----------
action : array_like
A 1-D array of actions taken by the AST Solver which deterministically control
a single step forward in the simulation.
Returns
-------
array_like
An observation from the timestep, determined by the settings and the `observation_return` helper function.
"""
# return None
# get the action from the list
self._action = action
# move the peds
self.update_peds()
# move the car
self._car = self.move_car(self._car, self._car_accel)
# take new measurements and noise them
noise = self._action.reshape((self.c_num_peds, 6))[:, 2:6]
self._measurements = self.sensors(self._peds, noise)
# filter out the noise with an alpha-beta tracker
self._car_obs = self.tracker(self._car_obs, self._measurements)
# select the SUT action for the next timestep
self._car_accel[0] = self.update_car(self._car_obs, self._car[0])
# grab simulation state, if interactive
self.observe()
self.observation = np.ndarray.flatten(self._env_obs)
# record step variables
self.log()
return self.observation
def reset(self, s_0):
"""Resets the state of the environment, returning an initial observation.
Parameters
----------
s_0 : array_like
The initial conditions to reset the simulator to.
Returns
-------
array_like
An observation from the timestep, determined by the settings and the `observation_return` helper function.
"""
# initialize variables
self._info = []
self._step = 0
self._path_length = 0
self._is_terminal = False
self.initial_conditions = s_0
self._action = np.array([0] * (6 * self.c_num_peds))
self._first_step = True
# Get v_des if it is sampled from a range
v_des = self.initial_conditions[3 * self.c_num_peds]
# initialize SUT location
car_init_x = self.initial_conditions[3 * self.c_num_peds + 1]
self._car = np.array([v_des, 0.0, car_init_x, self.c_car_init_y])
# zero out the first SUT acceleration
self._car_accel = np.zeros((2))
# initialize pedestrian locations and velocities
pos = self.initial_conditions[0:2 * self.c_num_peds]
self.x = pos[0:self.c_num_peds * 2:2]
self.y = pos[1:self.c_num_peds * 2:2]
v_start = self.initial_conditions[2 * self.c_num_peds:3 * self.c_num_peds]
self._peds[0:self.c_num_peds, 0] = np.zeros((self.c_num_peds))
self._peds[0:self.c_num_peds, 1] = v_start
self._peds[0:self.c_num_peds, 2] = self.x
self._peds[0:self.c_num_peds, 3] = self.y
# Calculate the relative position measurements
self._measurements = self._peds
self._env_obs = self._measurements
self._car_obs = self._measurements
# return the initial simulation state
self.observation = np.ndarray.flatten(self._measurements)
# self.observation = obs
return self.observation
def collision_detected(self):
"""
Returns whether the current state is in the goal set.
Checks to see if any pedestrian's position violates both the `min_dist_x` and `min_dist_y` constraints.
Returns
-------
bool
True if current state is in goal set.
"""
# calculate the relative distances between the pedestrians and the car
dist = self._peds[:, 2:4] - self._car[2:4]
# return True if any relative distance is within the SUT's hitbox and the car is still moving
if (np.any(np.all(np.less_equal(abs(dist), self.c_min_dist), axis=1)) and
self._car[0] > 0.5):
return True
return False
def log(self):
"""
Perform any logging steps.
"""
# Create a cache of step specific variables for post-simulation analysis
cache = np.hstack([0.0, # Dummy, will be filled in with trial # during post processing in save_trials.py
self._step,
np.ndarray.flatten(self._car),
np.ndarray.flatten(self._peds),
np.ndarray.flatten(self._action),
np.ndarray.flatten(self._car_obs),
0.0])
self._info.append(cache)
self._step += 1
def sensors(self, peds, noise):
"""Get a noisy observation of the pedestrians' locations and velocities.
Parameters
----------
peds : array_like
Positions and velocities of the pedestrians.
noise : array_like
Noise to add to the positions and velocities of the pedestrians.
Returns
-------
array_like
Noisy observation of the pedestrians' locations and velocities.
"""
measurements = peds + noise
return measurements
def tracker(self, estimate_old, measurements):
"""An alpha-beta filter to smooth noisy observations into an estimate of pedestrian state.
Parameters
----------
estimate_old : array_like
The smoothed state estimate from the previous timestep.
measurements : array_like
The noisy observation of pedestrian state from the current timestep.
Returns
-------
array_like
The smoothed state estimate of pedestrian state from the current timestep.
"""
observation = np.zeros_like(estimate_old)
observation[:, 0:2] = estimate_old[:, 0:2]
observation[:, 2:4] = estimate_old[:, 2:4] + self.c_dt * estimate_old[:, 0:2]
residuals = measurements[:, 2:4] - observation[:, 2:4]
observation[:, 2:4] += self.c_alpha * residuals
observation[:, 0:2] += self.c_beta / self.c_dt * residuals
return observation
def update_car(self, obs, v_car):
"""Calculate the ego vehicle's acceleration.
Parameters
----------
obs : array_like
Smoothed estimate of pedestrian state from the `tracker`.
v_car : float
Current velocity of the ego vehicle.
Returns
-------
float
The acceleration of the ego vehicle.
"""
cond = np.repeat(np.resize(np.logical_and(obs[:, 3] > -1.5, obs[:, 3] < 4.5), (self.c_num_peds, 1)), 4, axis=1)
in_road = np.expand_dims(np.extract(cond, obs), axis=0)
if in_road.size != 0:
mins = np.argmin(in_road.reshape((-1, 4)), axis=0)
v_oth = obs[mins[3], 0]
s_headway = obs[mins[3], 2] - self._car[2]
s_headway = max(10 ** -6, abs(s_headway)) * np.sign(s_headway) # avoid div by zero error later
del_v = v_oth - v_car
s_des = self.c_s_min + v_car * self.c_t_headway - v_car * del_v / (2 * np.sqrt(self.c_a_max * self.c_d_cmf))
if self.c_v_des > 0.0:
v_ratio = v_car / self.c_v_des
else:
v_ratio = 1.0
a = self.c_a_max * (1.0 - v_ratio ** self.c_delta - (s_des / s_headway) ** 2)
else:
del_v = self.c_v_des - v_car
a = del_v
if np.isnan(a):
pdb.set_trace()
# pdb.set_trace()
return np.clip(a, -self.c_d_max, self.c_a_max)
def move_car(self, car, accel):
"""Update the ego vehicle's state.
Parameters
----------
car : array_like
The ego vehicle's state: [x-velocity, y-velocity, x-position, y-position].
accel : float
The ago vehicle's acceleration.
Returns
-------
array_like
An updated version of the ego vehicle's state.
"""
car[2:4] += self.c_dt * car[0:2]
car[0:2] += self.c_dt * accel
return car
def update_peds(self):
"""Update the pedestrian's state.
"""
# Update ped state from actions
action = self._action.reshape((self.c_num_peds, 6))[:, 0:2]
mod_a = np.hstack((action,
self._peds[:, 0:2] + 0.5 * self.c_dt * action))
if np.any(np.isnan(mod_a)):
pdb.set_trace()
self._peds += self.c_dt * mod_a
# Enforce max abs(velocity) on pedestrians
self._peds[:, 0:2] = np.clip(self._peds[:, 0:2], a_min=[-4.5, -4.5], a_max=[4.5, 4.5])
if np.any(np.isnan(self._peds)):
pdb.set_trace()
def observe(self):
"""Get the ground truth state of the pedestrian relative to the ego vehicle.
"""
self._env_obs = self._peds - self._car
def get_ground_truth(self):
"""Clones the ground truth simulator state.
Returns
-------
dict
A dictionary of simulator state variables.
"""
return {'step': self._step,
'path_length': self._path_length,
'is_terminal': self._is_terminal,
'car': self._car,
'car_accel': self._car_accel,
'peds': self._peds,
'car_obs': self._car_obs,
'action': self._action,
'initial_conditions': self.initial_conditions,
}
def set_ground_truth(self, in_simulator_state):
"""Sets the simulator state variables.
Parameters
----------
in_simulator_state : dict
A dictionary of simulator state variables.
"""
in_simulator_state.copy()
self._step = in_simulator_state['step']
self._path_length = in_simulator_state['path_length']
self._is_terminal = in_simulator_state['is_terminal']
self._car = in_simulator_state['car']
self._car_accel = in_simulator_state['car_accel']
self._peds = in_simulator_state['peds']
self._car_obs = in_simulator_state['car_obs']
self._action = in_simulator_state['action']
self.initial_conditions = np.array(in_simulator_state['initial_conditions'])
self.observe()
self.observation = self._env_obs