PythonLinearNonlinearControl/PythonLinearNonlinearControl/controllers/mppi_williams.py

146 lines
4.7 KiB
Python

from logging import getLogger
import numpy as np
import scipy.stats as stats
from .controller import Controller
from ..envs.cost import calc_cost
logger = getLogger(__name__)
class MPPIWilliams(Controller):
""" Model Predictive Path Integral for linear and nonlinear method
Attributes:
history_u (list[numpy.ndarray]): time history of optimal input
Ref:
G. Williams et al., "Information theoretic MPC
for model-based reinforcement learning,"
2017 IEEE International Conference on Robotics and Automation (ICRA),
Singapore, 2017, pp. 1714-1721.
"""
def __init__(self, config, model):
super(MPPIWilliams, self).__init__(config, model)
# model
self.model = model
# general parameters
self.pred_len = config.PRED_LEN
self.input_size = config.INPUT_SIZE
# mppi parameters
self.pop_size = config.opt_config["MPPIWilliams"]["popsize"]
self.lam = config.opt_config["MPPIWilliams"]["lambda"]
self.noise_sigma = config.opt_config["MPPIWilliams"]["noise_sigma"]
self.opt_dim = self.input_size * self.pred_len
# get bound
self.input_upper_bounds = np.tile(config.INPUT_UPPER_BOUND,
(self.pred_len, 1))
self.input_lower_bounds = np.tile(config.INPUT_LOWER_BOUND,
(self.pred_len, 1))
# get cost func
self.state_cost_fn = config.state_cost_fn
self.terminal_state_cost_fn = config.terminal_state_cost_fn
self.input_cost_fn = config.input_cost_fn
# init mean
self.prev_sol = np.tile((config.INPUT_UPPER_BOUND
+ config.INPUT_LOWER_BOUND) / 2.,
self.pred_len)
self.prev_sol = self.prev_sol.reshape(self.pred_len, self.input_size)
# save
self.history_u = [np.zeros(self.input_size)]
def clear_sol(self):
""" clear prev sol
"""
logger.debug("Clear Solution")
self.prev_sol = \
(self.input_upper_bounds + self.input_lower_bounds) / 2.
self.prev_sol = self.prev_sol.reshape(self.pred_len, self.input_size)
def calc_cost(self, curr_x, samples, g_xs):
""" calculate the cost of input samples by using MPPI's eq
Args:
curr_x (numpy.ndarray): shape(state_size),
current robot position
samples (numpy.ndarray): shape(pop_size, opt_dim),
input samples
g_xs (numpy.ndarray): shape(pred_len, state_size),
goal states
Returns:
costs (numpy.ndarray): shape(pop_size, )
"""
# get size
pop_size = samples.shape[0]
g_xs = np.tile(g_xs, (pop_size, 1, 1))
# calc cost, pred_xs.shape = (pop_size, pred_len+1, state_size)
pred_xs = self.model.predict_traj(curr_x, samples)
# get particle cost
costs = calc_cost(pred_xs, samples, g_xs,
self.state_cost_fn, None,
self.terminal_state_cost_fn)
return costs
def obtain_sol(self, curr_x, g_xs):
""" calculate the optimal inputs
Args:
curr_x (numpy.ndarray): current state, shape(state_size, )
g_xs (numpy.ndarrya): goal trajectory, shape(plan_len, state_size)
Returns:
opt_input (numpy.ndarray): optimal input, shape(input_size, )
"""
# get noised inputs
noise = np.random.normal(
loc=0, scale=1.0, size=(self.pop_size, self.pred_len,
self.input_size)) * self.noise_sigma
noised_inputs = self.prev_sol + noise
# clip actions
noised_inputs = np.clip(
noised_inputs, self.input_lower_bounds, self.input_upper_bounds)
# calc cost
costs = self.calc_cost(curr_x, noised_inputs, g_xs)
costs += np.sum(np.sum(
self.lam * self.prev_sol * noise / self.noise_sigma,
axis=-1), axis=-1)
# mppi update
beta = np.min(costs)
eta = np.sum(np.exp(- 1. / self.lam * (costs - beta)), axis=0) \
+ 1e-10
# weight
# eta.shape = (pred_len, input_size)
weights = np.exp(- 1. / self.lam * (costs - beta)) / eta
# update inputs
sol = self.prev_sol \
+ np.sum(weights[:, np.newaxis, np.newaxis] * noise, axis=0)
# update
self.prev_sol[:-1] = sol[1:]
self.prev_sol[-1] = sol[-1] # last use the terminal input
# log
self.history_u.append(sol[0])
return sol[0]
def __str__(self):
return "MPPIWilliams"