Python load_mat示例

示例#1

文件： rl_control_sim_0423_mpc.py 项目： sbpac-office/pysim

from keras.models import Sequential, load_model

colors = dict(mcolors.BASE_COLORS, **mcolors.CSS4_COLORS)
app_dir = os.path.abspath('')
os.chdir('..\..\..')
from pysim.models.model_power import Mod_Power, Mod_Battery, Mod_Motor
from pysim.models.model_vehicle import Mod_Body, Mod_Veh
from pysim.models.model_environment import Mod_Env
from pysim.models.model_maneuver import Mod_Driver, Mod_Behavior
from pysim.sub_util.sub_type_def import type_DataLog, type_pid_controller
from pysim.sub_util.sub_utilities import Filt_LowPass
os.chdir(app_dir)
# import application modules
import get_data
get_data.set_dir(os.path.abspath('.\driving_data'))
DrivingData = get_data.load_mat('CfData1.mat')

get_data.set_dir(os.path.abspath('.\driver_data'))
DriverDataRaw = get_data.load_mat('Results_Sils_Learning.mat')
DriverDataRaw['BaseMap_InitDis'] = np.transpose(
    DriverDataRaw['BaseMap_InitDis'])
DriverDataRaw['BaseMap_CoastDis'] = np.transpose(
    DriverDataRaw['BaseMap_CoastDis'])

DriverDataKh = copy.deepcopy(DriverDataRaw)
DriverDataKh['LrnVec_Param_AccSlopeCf'] = np.expand_dims(np.transpose(
    DriverDataKh['MapArry_AccSlopeCf'][0, :]),
                                                         axis=1)
DriverDataKh['LrnVec_Param_RelDisAdj'] = np.expand_dims(np.transpose(
    DriverDataKh['MapArry_AdjDis'][0, :]),
                                                        axis=1)

示例#2

from keras.models import Sequential, load_model

colors = dict(mcolors.BASE_COLORS, **mcolors.CSS4_COLORS)
app_dir = os.path.abspath('')
os.chdir('..\..\..')
from pysim.models.model_power import Mod_Power
from pysim.models.model_vehicle import Mod_Body, Mod_Veh
from pysim.models.model_environment import Mod_Env
from pysim.models.model_maneuver import Mod_Driver, Mod_Behavior
from pysim.sub_util.sub_type_def import type_DataLog, type_pid_controller
from pysim.sub_util.sub_utilities import Filt_LowPass
os.chdir(app_dir)
# import application modules
import get_data
get_data.set_dir(os.path.abspath('.\driving_data'))
DrivingData = get_data.load_mat('CfData1.mat')

get_data.set_dir(os.path.abspath('.\driver_data'))
DriverData = get_data.load_mat('LrnVecCf.mat')

get_data.set_dir(os.path.abspath('.\model_data'))
kona_param = get_data.load_mat('CoeffSet_Kona.mat')
kona_param_est = get_data.load_mat('CoeffEst_Kona.mat')
Ts = 0.01

from rl_idm import IdmAccCf, DecelStateRecog, IdmClassic
from rl_environment import EnvRegen
from rl_algorithm import DdqrnAgent
from rl_etc_fcn import MovAvgFilt, fcn_plot_lrn_result, fcn_set_vehicle_param, fcn_log_data_store

示例#3

from pysim.models.model_vehicle import Mod_Body, Mod_Veh
from pysim.models.model_environment import Mod_Env
from pysim.models.model_maneuver import Mod_Driver, Mod_Behavior
from pysim.sub_util.sub_type_def import type_DataLog, type_pid_controller
from pysim.sub_util.sub_utilities import Filt_LowPass
os.chdir(app_dir)
# import application modules
import get_data
from rl_idm import IdmAccCf, DecelStateRecog, IdmClassic
from rl_environment import EnvRegen
from rl_algorithm import DdqrnAgent
from rl_etc_fcn import MovAvgFilt, fcn_plot_lrn_result, fcn_set_vehicle_param, fcn_log_data_store, fcn_driving_data_arrange, fcn_epdata_arrange
#%% import driving data
plt.close('all')
get_data.set_dir(os.path.abspath('.\driver_data'))
DriverData = get_data.load_mat('LrnVecCf.mat')

get_data.set_dir(os.path.abspath('.\model_data'))
kona_param = get_data.load_mat('CoeffSet_Kona.mat')
kona_param_est = get_data.load_mat('CoeffEst_Kona.mat')

Ts = 0.01

#%% MPC controller

# Discrete time model of a quadcopter
Ad = sparse.csc_matrix([[1, 0.01], [0, 1]])
Bd = sparse.csc_matrix([[-0.01 * 0.01 / 2], [-0.01]])

[nx, nu] = Bd.shape

示例#4

文件： rl_idm.py 项目： sbpac-office/pysim

* kyunghan <*****@*****.**>

Description
~~~~~~~~~~~~~
* Intelligent driver model

Update
~~~~~~~~~~~~~
* [19/02/22] - Initial draft design
"""
import numpy as np
import math
import os
import get_data
get_data.set_dir(os.path.abspath('.\driver_data'))
EffectProbIndex = get_data.load_mat('EffectiveProbIndex.mat')['ParLrn_ProbVec']


def EffectiveProbability(Index, IndexVectorArray, Std):
    "Calculation of effective probability"
    CoastIndexLength = np.size(IndexVectorArray)
    ProbVec = np.arange(CoastIndexLength, dtype='float')
    # Gaussian distribution
    for i in range(CoastIndexLength):
        VecVal = 1 / (Std * math.sqrt(2 * math.pi)) * np.exp(-0.5 * (
            (Index - IndexVectorArray[i]) / Std)**2)
        ProbVec[i, ] = VecVal[0, 0]
    # Normalization
    ProbSum = np.sum(ProbVec)
    EffProb = ProbVec / ProbSum
    return EffProb