Source code for blackbox_mpc.optimizers.pi2

import tensorflow as tf
import numpy as np
import tensorflow_probability as tfp
from blackbox_mpc.optimizers.optimizer_base import OptimizerBase
tfd = tfp.distributions


[docs]class PI2Optimizer(OptimizerBase):
[docs] def __init__(self, env_action_space, env_observation_space, planning_horizon=50, max_iterations=5, population_size=500, num_agents=5, lamda=tf.constant(1.0, dtype=tf.float32)): """ This class defines the information theortic MPC based on path intergral methods. (https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7989202) Parameters --------- env_action_space: gym.ActionSpace Defines the action space of the gym environment. env_observation_space: gym.ObservationSpace Defines the observation space of the gym environment. planning_horizon: Int Defines the planning horizon for the optimizer (how many steps to lookahead and optimize for). max_iterations: tf.int32 Defines the maximimum iterations for the CMAES optimizer to refine its guess for the optimal solution. population_size: tf.int32 Defines the population size of the particles evaluated at each iteration. num_agents: tf.int32 Defines the number of runner running in parallel lamda: tf.float32 Defines the lamda used the energy function. """ super(PI2Optimizer, self).__init__(name=None, planning_horizon=planning_horizon, max_iterations=max_iterations, num_agents=num_agents, env_action_space=env_action_space, env_observation_space= env_observation_space) self._solution_dim = [self._num_agents, self._planning_horizon, self._dim_U] self._solution_size = tf.reduce_prod(self._solution_dim) self._population_size = population_size previous_solution_values = np.tile((self._action_lower_bound + self._action_upper_bound) / 2, [self._planning_horizon * self._num_agents, 1]) previous_solution_values = previous_solution_values.reshape([self._num_agents, self._planning_horizon, -1]) self._previous_solution = tf.Variable(tf.zeros(shape=previous_solution_values.shape, dtype=tf.float32)) self._previous_solution.assign(previous_solution_values) solution_variance_values = np.tile(np.square(self._action_lower_bound - self._action_upper_bound) / 16, [self._planning_horizon * self._num_agents, 1]) solution_variance_values = solution_variance_values.reshape([self._num_agents, self._planning_horizon, -1]) self._solution_variance = tf.Variable(tf.zeros(shape=solution_variance_values.shape, dtype=tf.float32)) self._solution_variance.assign(solution_variance_values) self._lamda = lamda
@tf.function def _optimize(self, current_state, time_step): def continue_condition(t, mean): result = tf.less(t, self._max_iterations) return result def iterate(t, mean): samples = tf.random.truncated_normal([self._population_size, *self._solution_dim], mean, tf.sqrt(self._solution_variance), dtype=tf.float32) samples_feasible = tf.clip_by_value(samples, self._action_lower_bound_horizon, self._action_upper_bound_horizon) penalty = tf.norm(tf.reshape(samples - samples_feasible, [self._population_size, self._num_agents, -1]), axis=2) ** 2 samples = samples_feasible rewards = self._trajectory_evaluator(current_state, samples, time_step) - penalty costs = -rewards costs = tf.transpose(costs, [1, 0]) beta = tf.reduce_min(costs, axis=1) prob = tf.math.exp(-(1 / self._lamda) * (costs - tf.expand_dims(beta, -1))) eta = tf.reduce_sum(prob, axis=1) #compute weights now omega = tf.expand_dims(1 / eta, -1) * prob samples = tf.transpose(samples, [1, 0, 2, 3]) new_mean = tf.reduce_sum(tf.multiply(samples, tf.expand_dims(tf.expand_dims(omega, -1), -1)), axis=1) return t + tf.constant(1, dtype=tf.int32), new_mean _, new_mean = tf.while_loop(cond=continue_condition, body=iterate, loop_vars=[tf.constant(0, dtype=tf.int32), self._previous_solution]) self._previous_solution.assign(tf.concat([new_mean[:, 1:], tf.expand_dims(new_mean[:, -1], 1)], 1)) resulting_action = new_mean[:, 0] return resulting_action
[docs] def reset(self): """ This method resets the optimizer to its default state at the beginning of the trajectory/episode. """ previous_solution_values = np.tile((self._action_lower_bound + self._action_upper_bound) / 2, [self._planning_horizon * self._num_agents, 1]) previous_solution_values = previous_solution_values.reshape([self._num_agents, self._planning_horizon, -1]) self._previous_solution.assign(previous_solution_values)