InformedStateSampler.cpp

/*********************************************************************
 * Software License Agreement (BSD License)
 *
 *  Copyright (c) 2014, University of Toronto
 *  All rights reserved.
 *
 *  Redistribution and use in source and binary forms, with or without
 *  modification, are permitted provided that the following conditions
 *  are met:
 *
 *   * Redistributions of source code must retain the above copyright
 *     notice, this list of conditions and the following disclaimer.
 *   * Redistributions in binary form must reproduce the above
 *     copyright notice, this list of conditions and the following
 *     disclaimer in the documentation and/or other materials provided
 *     with the distribution.
 *   * Neither the name of the University of Toronto nor the names of its
 *     contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission.
 *
 *  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 *  "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 *  LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
 *  FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
 *  COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
 *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
 *  BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
 *  LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
 *  CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 *  LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
 *  ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 *  POSSIBILITY OF SUCH DAMAGE.
 *********************************************************************/
 
/* Authors: Jonathan Gammell */
 
#include "ompl/base/samplers/InformedStateSampler.h"
#include "ompl/util/Exception.h"
#include "ompl/base/OptimizationObjective.h"
// The goal definitions
#include "ompl/base/Goal.h"
 
namespace ompl
{
    namespace base
    {
        // InformedSampler
        InformedSampler::InformedSampler(const ProblemDefinitionPtr &probDefn, unsigned int maxNumberCalls)
          : probDefn_(probDefn), space_(probDefn->getSpaceInformation()->getStateSpace()), numIters_(maxNumberCalls)
        {
            // Sanity check the problem.
            // Check that there is an optimization objective
            if (!probDefn_->hasOptimizationObjective())
            {
                throw Exception("InformedSampler: An optimization objective must be specified at construction.");
            }
            // No else
 
            // Make sure we have at least one start and warn if we have more than one
            if (probDefn_->getStartStateCount() == 0u)
            {
                throw Exception("InformedSampler: At least one start state must be specified at construction.");
            }
            // No else
 
            // Store the optimization objective for later ease.
            opt_ = probDefn_->getOptimizationObjective();
        }
 
        double InformedSampler::getInformedMeasure(const Cost &minCost, const Cost &maxCost) const
        {
            // Subtract the measures defined by the max and min costs. These will be defined in the deriving class.
            return getInformedMeasure(maxCost) - getInformedMeasure(minCost);
        }
 
        Cost InformedSampler::heuristicSolnCost(const State *statePtr) const
        {
            // Return the best heuristic estimate of the cost-to-come and cost-to-go from the state considering all
            // starts.
 
            // If there's only one start, be simple:
            if (probDefn_->getStartStateCount() == 1u)
            {
                // Calculate and from the one and only start
                return opt_->combineCosts(opt_->motionCostHeuristic(probDefn_->getStartState(0u), statePtr),
                                          opt_->costToGo(statePtr, probDefn_->getGoal().get()));
            }
 
            // Calculate and return the best
 
            // Variable
            // The best cost so far
            Cost bestCost = opt_->infiniteCost();
 
            // Iterate over each start and store the best
            for (unsigned int i = 0u; i < probDefn_->getStartStateCount(); ++i)
            {
                // Store the best
                bestCost = opt_->betterCost(
                    bestCost, opt_->combineCosts(opt_->motionCostHeuristic(probDefn_->getStartState(i), statePtr),
                                                 opt_->costToGo(statePtr, probDefn_->getGoal().get())));
            }
 
            // Return the best
            return bestCost;
        }
 
        ProblemDefinitionPtr InformedSampler::getProblemDefn() const
        {
            return probDefn_;
        }
 
        unsigned int InformedSampler::getMaxNumberOfIters() const
        {
            return numIters_;
        }


        // InformedStateSampler
        InformedStateSampler::InformedStateSampler(const ProblemDefinitionPtr &probDefn, unsigned int maxNumberCalls,
                                                   const GetCurrentCostFunc &costFunc)
          : StateSampler(probDefn->getSpaceInformation()->getStateSpace().get())
        {
            // Call the common constructor with the default informed sampler
            commonConstructor(
                costFunc, probDefn->getOptimizationObjective()->allocInformedStateSampler(probDefn, maxNumberCalls));
        }
 
        InformedStateSampler::InformedStateSampler(const ProblemDefinitionPtr &probDefn,
                                                   const GetCurrentCostFunc &costFunc,
                                                   const InformedSamplerPtr &infSampler)
          : StateSampler(probDefn->getSpaceInformation()->getStateSpace().get())
        {
            // Call the common constructor with the given informed sampler
            commonConstructor(costFunc, infSampler);
        }
 
        void InformedStateSampler::commonConstructor(const GetCurrentCostFunc &costFunc,
                                                     const InformedSamplerPtr &infSampler)
        {
            // Store the cost function
            bestCostFunc_ = costFunc;
 
            // Store the informed sampler
            infSampler_ = infSampler;
 
            // Allocate a base sampler
            baseSampler_ = StateSampler::space_->allocDefaultStateSampler();
        }
 
        void InformedStateSampler::sampleUniform(State *statePtr)
        {
            // Variable
            // Whether informed sampling was successful
            bool informedSuccess;
 
            // Call sample uniform with the current best cost, check returning function:
            informedSuccess = infSampler_->sampleUniform(statePtr, bestCostFunc_());
 
            // If we were unsuccessful, return a regular sample
            if (!informedSuccess)
            {
                baseSampler_->sampleUniform(statePtr);
            }
            // No else.
        }
 
        void InformedStateSampler::sampleUniformNear(State *statePtr, const State *near, const double distance)
        {
            // Warn:
            OMPL_WARN("sampleUniformNear is not informed.");
            return baseSampler_->sampleUniformNear(statePtr, near, distance);
        }
 
        void InformedStateSampler::sampleGaussian(State *statePtr, const State *mean, const double stdDev)
        {
            // Warn:
            OMPL_WARN("sampleGaussian is not informed.");
            return baseSampler_->sampleGaussian(statePtr, mean, stdDev);
        }

    };  // namespace base
};  // namespace ompl