def test1(self):
global test_file
hdf5file = test_file[:-4]+'.hdf5'
elems_indxs = [('CFS_Accelerator_Pedal_Position', 0),
('SCC_Spline_Lane_Deviation', 1),
('SCC_Spline_Lane_Deviation', 3),
('VDS_Tire_Weight_On_Wheels', slice(0,4))]
daq = Daq()
daq.read_hd5(os.path.join('data', hdf5file))
# print(daq['VDS_Tire_Weight_On_Wheels'].frames.shape)
fig = daq.plot_ts(elems_indxs, fslice(6000, None))
fig.savefig('./output/daq_plots_test.png')
def test3(self):
x = self.x
rs = np.array([[24,245,6325,2435,245,1234,14,234,548]])
ds = x[:, fslice(3000, 3009)]
assert_array_equal(ds, rs)
def test2(self):
x = self.x
rs = np.array([[24,6325,245,14,548]])
assert_array_equal(x[:, fslice(None, None, 2)], rs)
def test1(self):
x = self.x
rs = np.array([[ 245., 6325., 2435.]])
assert_array_equal(x[:, fslice(3001, 3004)], rs)
def test00(self):
x = self.x
rs = np.array([3000, 3001, 3002, 3003])
assert_array_equal(x[:, fslice(3004)].frames, rs)
i*10+3 in daq.etc['epochs']):
# encountered partial trial
continue
scenario = daq.etc['scen_order'][i]
# unpack the start frame of the 1-to-2 lane addition transition
# and the stop frame of the 2-to-1 lane reduction transition
f0 = daq.etc['epochs'][i*10+1].start
fend = daq.etc['epochs'][i*10+3].stop
print(" on scenario %i [f%i:f%i]..."%(scenario, f0, fend))
# next we need to linearly interpolate speed by distance
veh_dist = daq['VDS_Veh_Dist'][0, fslice(f0, fend)].flatten()
veh_dist -= veh_dist[0]
veh_spd = daq['VDS_Veh_Speed'][0, fslice(f0, fend)].flatten()
veh_spd_ip = np.interp(veh_dist_ip, veh_dist, veh_spd)
# store interpolated speed so we can plot it
interp_spds_by_scenario[scenario].append(veh_spd_ip)
print('\nStep 2. Plot speeds...')
# names of scenarios
scenario_names = ['Baseline',
'Advisory',
'Reg',
'Reg + Adv',
i*10+3 in daq.etc['epochs']):
# encountered partial trial
continue
scenario = daq.etc['scen_order'][i]
# unpack the start frame of the 1-to-2 lane addition transition
# and the stop frame of the 2-to-1 lane reduction transition
f0 = daq.etc['epochs'][i * 10 + 1].start
fend = daq.etc['epochs'][i * 10 + 3].stop
print(" on scenario %i [f%i:f%i]..." % (scenario, f0, fend))
# next we need to linearly interpolate speed by distance
veh_dist = daq['VDS_Veh_Dist'][0, fslice(f0, fend)].flatten()
veh_dist -= veh_dist[0]
veh_spd = daq['VDS_Veh_Speed'][0, fslice(f0, fend)].flatten()
veh_spd_ip = np.interp(veh_dist_ip, veh_dist, veh_spd)
# store interpolated speed so we can plot it
interp_spds_by_scenario[scenario].append(veh_spd_ip)
print('\nStep 2. Plot speeds...')
# names of scenarios
scenario_names = [
'Baseline', 'Advisory', 'Reg', 'Reg + Adv', 'Chevrons', 'Lines',
'Narrowing', 'Parallax', 'Force Rh', 'Lines w/ mid'
]
def mult_ado_by_pid_plot(pids, page):
global hdf5_files, scenario_names, latin_square, get_pid
# initialize plot
fig = plt.figure(figsize=(11*1.75, 8.5*1.75))
fig.subplots_adjust(left=.04, right=.96,
bottom=.05, top=.97,
wspace=.04, hspace=.04)
# plot will have 2 axis. One is for Vehicle Speed, the other is
# for relative distance of the DynObjs
xticks = np.linspace(660,5280*1.25,10)
xlim = [0, 5280*1.25]
# speed in MPH
yticks1 = np.linspace(35,75,5)
ylim1 = [30,80]
# relative distance in Feet
yticks2 = np.linspace(-1500,1500,7)
ylim2 = [-1750,1750]
# loop through the hdf5_files that contain
# data for the requested participants
for hdf5_file in [hd for hd in hdf5_files if get_pid(hd) in pids]:
print('analyzing "%s"'%hdf5_file)
daq = Daq()
daq.read_hd5(hdf5_file)
pid = daq.etc['pid']
# figure out row number
rnum = pids.index(pid)
# find the relevant dynamic objects to plot
platoon = [do for do in daq.dynobjs.values() if 'Ado' in do.name]
platoon = sorted(platoon, key=attrgetter('name'))
# for each trial...
for i in xrange(10):
if not ( i*10+1 in daq.etc['epochs'] and \
i*10+3 in daq.etc['epochs']):
# encountered partial trial
continue
scenario = daq.etc['scen_order'][i]
scenario_name = scenario_names[scenario]
print(' PID: %03i, Passing zone: %i (%s)'%(pid, i, scenario_name))
# unpack the start frame of the 1-to-2 lane addition transition
# and the stop frame of the 2-to-1 lane reduction transition
f0 = daq.etc['epochs'][i*10+1].start
fend = daq.etc['epochs'][i*10+3].stop
# get axis handle
ax1 = plt.subplot(8, 10, rnum*10 + scenario + 1)
distance = daq['VDS_Veh_Dist'][0, fslice(f0, fend)].flatten()
distance -= distance[0]
speed = daq['VDS_Veh_Speed'][0, fslice(f0, fend)].flatten()
ax1.plot(distance, speed, 'b')
ax2 = ax1.twinx()
# we need to figure out when the Ados enter the passing lane
# and exit the passing lane. To do this we need to know where
# the passing lane starts and ends
pos0 = daq['VDS_Chassis_CG_Position'][:,findex(f0)]
posend = daq['VDS_Chassis_CG_Position'][:,findex(fend)]
# loop through and plot ados
for j, do in enumerate(platoon):
# each passing zone has its own set of Ados. This
# sorts out which are actually defined (should be
# defined) for this trial.
if i*10 <= int(do.name[-2:]) < (i+1)*10:
print(do.name)
# figure out when the ado enters the
d0 = np.sum((do.pos - pos0)**2., axis=0)**.5
dend = np.sum((do.pos - posend)**2., axis=0)**.5
# indexes relative to do arrays
imin0 = d0.argmin()
iminend = dend.argmin()
# now we can plot the Ados relative distance to the
# vehicle as a function of distance over the passing
# lane
if iminend-imin0 > 0:
distance = do.distance[0, imin0:iminend]
distance -= distance[0]
rel_distance = do.relative_distance[0, imin0:iminend]
ax2.plot(distance, rel_distance, color='g', alpha=0.4)
else:
print(' %s did not drive '
'through passing zone'%do.name)
# make things pretty
ax1.set_ylim(ylim1)
ax1.set_yticks(yticks1)
if scenario:
ax1.set_yticklabels(['' for y in yticks1])
else:
ax1.set_yticklabels(yticks1, color='b')
ax2.axhline(0, color='k', linestyle=':')
ax2.set_ylim(ylim2)
ax2.set_yticks(yticks2)
if scenario != 9:
ax2.set_yticklabels(['' for y in yticks2])
else:
ax2.set_yticklabels(yticks2, color='g')
ax2.set_xlim(xlim)
ax2.set_xticks(xticks)
if rnum == 0:
ax1.set_title(scenario_names[scenario])
if rnum == len(pids)-1:
ax1.set_xticklabels(['%i'%x for x in xticks],
rotation='vertical')
else:
ax1.set_xticklabels(['' for x in xticks])
if not scenario:
ax1.text(660,35,'Participant: %03i =>'%pids[rnum],
size='small')
img_name = 'do_passing_behavior__PAGE%i.png'%page
fig.savefig(img_name, dpi=300)
plt.close()
def test3(self):
x = self.x
rs = np.array([[24, 245, 6325, 2435, 245, 1234, 14, 234, 548]])
ds = x[:, fslice(3000, 3009)]
assert_array_equal(ds, rs)
def test2(self):
x = self.x
rs = np.array([[24, 6325, 245, 14, 548]])
assert_array_equal(x[:, fslice(None, None, 2)], rs)
def test1(self):
x = self.x
rs = np.array([[245., 6325., 2435.]])
assert_array_equal(x[:, fslice(3001, 3004)], rs)
def test00(self):
x = self.x
rs = np.array([3000, 3001, 3002, 3003])
assert_array_equal(x[:, fslice(3004)].frames, rs)
