RRTXstatic.h

/*********************************************************************
* Software License Agreement (BSD License)
*
*  Copyright (c) 2016, Georgia Institute of Technology
*  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 Rice University 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.
*********************************************************************/
 
/* Author: Florian Hauer */
 
#ifndef OMPL_GEOMETRIC_PLANNERS_RRT_RRTXSTATIC_
#define OMPL_GEOMETRIC_PLANNERS_RRT_RRTXSTATIC_
 
#include <ompl/datastructures/BinaryHeap.h>
#include "ompl/base/OptimizationObjective.h"
#include "ompl/datastructures/NearestNeighbors.h"
#include "ompl/geometric/planners/PlannerIncludes.h"
 
#include <deque>
#include <limits>
#include <list>
#include <queue>
#include <utility>
#include <vector>
 
namespace ompl
{
    namespace geometric
    {
        class RRTXstatic : public base::Planner
        {
        public:
            RRTXstatic(const base::SpaceInformationPtr &si);
 
            ~RRTXstatic() override;
 
            void getPlannerData(base::PlannerData &data) const override;
 
            base::PlannerStatus solve(const base::PlannerTerminationCondition &ptc) override;
 
            void clear() override;
 
            void setup() override;
 
            void setGoalBias(double goalBias)
            {
                goalBias_ = goalBias;
            }
 
            double getGoalBias() const
            {
                return goalBias_;
            }
 
            void setInformedSampling(bool informedSampling);
 
            bool getInformedSampling() const
            {
                return useInformedSampling_;
            }
 
            void setSampleRejection(bool reject);
 
            bool getSampleRejection() const
            {
                return useRejectionSampling_;
            }
 
            void setNumSamplingAttempts(unsigned int numAttempts)
            {
                numSampleAttempts_ = numAttempts;
            }
 
            unsigned int getNumSamplingAttempts() const
            {
                return numSampleAttempts_;
            }
 
            virtual void setEpsilon(double epsilon)
            {
                epsilonCost_ = base::Cost(epsilon);
            }
 
            double getEpsilon() const
            {
                return epsilonCost_.value();
            }
 
            void setRange(double distance)
            {
                maxDistance_ = distance;
            }
 
            double getRange() const
            {
                return maxDistance_;
            }
 
            void setRewireFactor(double rewireFactor)
            {
                rewireFactor_ = rewireFactor;
                calculateRewiringLowerBounds();
            }
 
            double getRewireFactor() const
            {
                return rewireFactor_;
            }
 
            template <template <typename T> class NN>
            void setNearestNeighbors()
            {
                if (nn_ && nn_->size() != 0)
                    OMPL_WARN("Calling setNearestNeighbors will clear all states.");
                clear();
                nn_ = std::make_shared<NN<Motion *>>();
                setup();
            }
 
            void setKNearest(bool useKNearest)
            {
                useKNearest_ = useKNearest;
            }
 
            bool getKNearest() const
            {
                return useKNearest_;
            }
 
            void setUpdateChildren(bool val)
            {
                updateChildren_ = val;
            }
 
            bool getUpdateChildren() const
            {
                return updateChildren_;
            }
 
            void setVariant(const int variant)
            {
                if (variant < 0 || variant > 3)
                    throw Exception("Variant must be 0 (original RRT#) or in [1, 3]");
                variant_ = variant;
            }
 
            int getVariant() const
            {
                return variant_;
            }
 
            void setAlpha(const double a)
            {
                alpha_ = a;
            }
 
            double getAlpha() const
            {
                return alpha_;
            }
 
            unsigned int numIterations() const
            {
                return iterations_;
            }
 
            ompl::base::Cost bestCost() const
            {
                return bestCost_;
            }
 
        protected:
            class Motion;
 
            struct MotionCompare
            {
                MotionCompare(base::OptimizationObjectivePtr opt, base::ProblemDefinitionPtr pdef)
                  : opt_(std::move(opt)), pdef_(std::move(pdef))
                {
                }
 
                inline base::Cost costPlusHeuristic(const Motion *m) const
                {
                    return opt_->combineCosts(m->cost, opt_->costToGo(m->state, pdef_->getGoal().get()));
                }
 
                inline base::Cost alphaCostPlusHeuristic(const Motion *m, double alpha) const
                {
                    return opt_->combineCosts(base::Cost(alpha * m->cost.value()),
                                              opt_->costToGo(m->state, pdef_->getGoal().get()));
                }
 
                inline bool operator()(const Motion *m1, const Motion *m2) const
                {
                    // we use a max heap, to do a min heap so the operator < returns > in order to make it a min heap
                    return !opt_->isCostBetterThan(costPlusHeuristic(m1), costPlusHeuristic(m2));
                }
 
                base::OptimizationObjectivePtr opt_;
 
                base::ProblemDefinitionPtr pdef_;
            };
 
            class Motion
            {
            public:
                Motion(const base::SpaceInformationPtr &si) : state(si->allocState()), parent(nullptr), handle(nullptr)
                {
                }
 
                ~Motion() = default;
 
                base::State *state;
 
                Motion *parent;
 
                base::Cost cost;
 
                std::vector<Motion *> children;
 
                std::vector<std::pair<Motion *, bool>> nbh;
 
                BinaryHeap<Motion *, MotionCompare>::Element *handle;
            };
 
            void allocSampler();
 
            bool sampleUniform(base::State *statePtr);
 
            void freeMemory();
 
            double distanceFunction(const Motion *a, const Motion *b) const
            {
                return si_->distance(a->state, b->state);
            }
 
            void updateQueue(Motion *x);
 
            void removeFromParent(Motion *m);
 
            void getNeighbors(Motion *motion) const;
 
            void calculateRewiringLowerBounds();
 
            void calculateRRG();
 
            bool includeVertex(const Motion *x) const;
 
            base::StateSamplerPtr sampler_;
 
            base::InformedSamplerPtr infSampler_;
 
            std::shared_ptr<NearestNeighbors<Motion *>> nn_;
 
            double goalBias_{.05};
 
            double maxDistance_{0.};
 
            RNG rng_;
 
            bool useKNearest_{true};
 
            double rewireFactor_{1.1};
 
            double k_rrt_{0u};
            double r_rrt_{0.};
 
            base::OptimizationObjectivePtr opt_;
 
            Motion *lastGoalMotion_{nullptr};
 
            std::vector<Motion *> goalMotions_;
 
            base::Cost bestCost_{std::numeric_limits<double>::quiet_NaN()};
 
            unsigned int iterations_{0u};
 
            MotionCompare mc_;
 
            BinaryHeap<Motion *, MotionCompare> q_;
 
            base::Cost epsilonCost_{0.};
 
            bool updateChildren_{true};
 
            double rrg_r_;
 
            unsigned int rrg_k_;
 
            int variant_{0};
 
            double alpha_{1.};
 
            bool useInformedSampling_{false};
 
            bool useRejectionSampling_{false};
 
            unsigned int numSampleAttempts_{100u};
 
            // Planner progress property functions
            std::string numIterationsProperty() const
            {
                return std::to_string(numIterations());
            }
            std::string bestCostProperty() const
            {
                return std::to_string(bestCost().value());
            }
            std::string numMotionsProperty() const
            {
                return std::to_string(nn_->size());
            }
        };
    }
}
 
#endif