Loading...
Searching...
No Matches
LBTRRT.cpp
1/*********************************************************************
2* Software License Agreement (BSD License)
3*
4* Copyright (c) 2015, Tel Aviv University
5* All rights reserved.
6*
7* Redistribution and use in source and binary forms, with or without
8* modification, are permitted provided that the following conditions
9* are met:
10*
11* * Redistributions of source code must retain the above copyright
12* notice, this list of conditions and the following disclaimer.
13* * Redistributions in binary form must reproduce the above
14* copyright notice, this list of conditions and the following
15* disclaimer in the documentation and/or other materials provided
16* with the distribution.
17* * Neither the name of the Tel Aviv University nor the names of its
18* contributors may be used to endorse or promote products derived
19* from this software without specific prior written permission.
20*
21* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
24* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
25* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
26* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
27* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
28* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
29* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
31* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
32* POSSIBILITY OF SUCH DAMAGE.
33*********************************************************************/
34
35/* Author: Oren Salzman, Aditya Mandalika, Sertac Karaman, Ioan Sucan, Mark Moll */
36
37#include "ompl/geometric/planners/rrt/LBTRRT.h"
38#include "ompl/base/goals/GoalSampleableRegion.h"
39#include "ompl/tools/config/SelfConfig.h"
40#include <limits>
41#include <cmath>
42#include <boost/math/constants/constants.hpp>
43
44ompl::geometric::LBTRRT::LBTRRT(const base::SpaceInformationPtr &si)
45 : base::Planner(si, "LBTRRT")
46{
48 specs_.directed = true;
49
50 Planner::declareParam<double>("range", this, &LBTRRT::setRange, &LBTRRT::getRange, "0.:1.:10000.");
51 Planner::declareParam<double>("goal_bias", this, &LBTRRT::setGoalBias, &LBTRRT::getGoalBias, "0.:.05:1.");
52 Planner::declareParam<double>("epsilon", this, &LBTRRT::setApproximationFactor, &LBTRRT::getApproximationFactor,
53 "0.:.1:10.");
54
55 addPlannerProgressProperty("iterations INTEGER", [this]
56 {
57 return getIterationCount();
58 });
59 addPlannerProgressProperty("best cost REAL", [this]
60 {
61 return getBestCost();
62 });
63}
64
65ompl::geometric::LBTRRT::~LBTRRT()
66{
67 freeMemory();
68}
69
71{
72 Planner::clear();
73 sampler_.reset();
74 freeMemory();
75 if (nn_)
76 nn_->clear();
77 lowerBoundGraph_.clear();
78 lastGoalMotion_ = nullptr;
79
80 iterations_ = 0;
81 bestCost_ = std::numeric_limits<double>::infinity();
82}
83
85{
86 Planner::setup();
87 tools::SelfConfig sc(si_, getName());
88 sc.configurePlannerRange(maxDistance_);
89
90 if (!nn_)
91 nn_.reset(tools::SelfConfig::getDefaultNearestNeighbors<Motion *>(this));
92 nn_->setDistanceFunction([this](const Motion *a, const Motion *b)
93 {
94 return distanceFunction(a, b);
95 });
96}
97
99{
100 if (!idToMotionMap_.empty())
101 {
102 for (auto &i : idToMotionMap_)
103 {
104 if (i->state_ != nullptr)
105 si_->freeState(i->state_);
106 delete i;
107 }
108 }
109 idToMotionMap_.clear();
110}
111
113{
114 checkValidity();
115 // update goal and check validity
116 base::Goal *goal = pdef_->getGoal().get();
117 auto *goal_s = dynamic_cast<base::GoalSampleableRegion *>(goal);
118
119 if (goal == nullptr)
120 {
121 OMPL_ERROR("%s: Goal undefined", getName().c_str());
123 }
124
125 // update start and check validity
126 while (const base::State *st = pis_.nextStart())
127 {
128 auto *motion = new Motion(si_);
129 si_->copyState(motion->state_, st);
130 motion->id_ = nn_->size();
131 idToMotionMap_.push_back(motion);
132 nn_->add(motion);
133 lowerBoundGraph_.addVertex(motion->id_);
134 }
135
136 if (nn_->size() == 0)
137 {
138 OMPL_ERROR("%s: There are no valid initial states!", getName().c_str());
140 }
141
142 if (nn_->size() > 1)
143 {
144 OMPL_ERROR("%s: There are multiple start states - currently not supported!", getName().c_str());
146 }
147
148 if (!sampler_)
149 sampler_ = si_->allocStateSampler();
150
151 OMPL_INFORM("%s: Starting planning with %u states already in datastructure", getName().c_str(), nn_->size());
152
153 Motion *solution = lastGoalMotion_;
154 Motion *approxSol = nullptr;
155 double approxdif = std::numeric_limits<double>::infinity();
156 // e*(1+1/d) K-nearest constant, as used in RRT*
157 double k_rrg =
158 boost::math::constants::e<double>() + boost::math::constants::e<double>() / (double)si_->getStateDimension();
159
160 auto *rmotion = new Motion(si_);
161 base::State *rstate = rmotion->state_;
162 base::State *xstate = si_->allocState();
163 unsigned int statesGenerated = 0;
164
165 bestCost_ = lastGoalMotion_ != nullptr ? lastGoalMotion_->costApx_ : std::numeric_limits<double>::infinity();
166 while (!ptc())
167 {
168 iterations_++;
169 /* sample random state (with goal biasing) */
170 if ((goal_s != nullptr) && rng_.uniform01() < goalBias_ && goal_s->canSample())
171 goal_s->sampleGoal(rstate);
172 else
173 sampler_->sampleUniform(rstate);
174
175 /* find closest state in the tree */
176 Motion *nmotion = nn_->nearest(rmotion);
177 base::State *dstate = rstate;
178
179 /* find state to add */
180 double d = si_->distance(nmotion->state_, rstate);
181 if (d == 0) // this takes care of the case that the goal is a single point and we re-sample it multiple times
182 continue;
183 if (d > maxDistance_)
184 {
185 si_->getStateSpace()->interpolate(nmotion->state_, rstate, maxDistance_ / d, xstate);
186 dstate = xstate;
187 }
188
189 if (checkMotion(nmotion->state_, dstate))
190 {
191 statesGenerated++;
192 /* create a motion */
193 auto *motion = new Motion(si_);
194 si_->copyState(motion->state_, dstate);
195
196 /* update fields */
197 double distN = distanceFunction(nmotion, motion);
198
199 motion->id_ = nn_->size();
200 idToMotionMap_.push_back(motion);
201 lowerBoundGraph_.addVertex(motion->id_);
202 motion->parentApx_ = nmotion;
203
204 std::list<std::size_t> dummy;
205 lowerBoundGraph_.addEdge(nmotion->id_, motion->id_, distN, false, dummy);
206
207 motion->costLb_ = nmotion->costLb_ + distN;
208 motion->costApx_ = nmotion->costApx_ + distN;
209 nmotion->childrenApx_.push_back(motion);
210
211 std::vector<Motion *> nnVec;
212 unsigned int k = std::ceil(k_rrg * log((double)(nn_->size() + 1)));
213 nn_->nearestK(motion, k, nnVec);
214 nn_->add(motion); // if we add the motion before the nearestK call, we will get ourselves...
215
216 IsLessThan isLessThan(this, motion);
217 std::sort(nnVec.begin(), nnVec.end(), isLessThan);
218
219 //-------------------------------------------------//
220 // Rewiring Part (i) - find best parent of motion //
221 //-------------------------------------------------//
222 if (motion->parentApx_ != nnVec.front())
223 {
224 for (auto potentialParent : nnVec)
225 {
226 double dist = distanceFunction(potentialParent, motion);
227 considerEdge(potentialParent, motion, dist);
228 }
229 }
230
231 //------------------------------------------------------------------//
232 // Rewiring Part (ii) //
233 // check if motion may be a better parent to one of its neighbors //
234 //------------------------------------------------------------------//
235 for (auto child : nnVec)
236 {
237 double dist = distanceFunction(motion, child);
238 considerEdge(motion, child, dist);
239 }
240
241 double dist = 0.0;
242 bool sat = goal->isSatisfied(motion->state_, &dist);
243
244 if (sat)
245 {
246 approxdif = dist;
247 solution = motion;
248 }
249 if (dist < approxdif)
250 {
251 approxdif = dist;
252 approxSol = motion;
253 }
254
255 if (solution != nullptr && bestCost_ != solution->costApx_)
256 {
257 OMPL_INFORM("%s: approximation cost = %g", getName().c_str(), solution->costApx_);
258 bestCost_ = solution->costApx_;
259 }
260 }
261 }
262
263 bool solved = false;
264 bool approximate = false;
265
266 if (solution == nullptr)
267 {
268 solution = approxSol;
269 approximate = true;
270 }
271
272 if (solution != nullptr)
273 {
274 lastGoalMotion_ = solution;
275
276 /* construct the solution path */
277 std::vector<Motion *> mpath;
278 while (solution != nullptr)
279 {
280 mpath.push_back(solution);
281 solution = solution->parentApx_;
282 }
283
284 /* set the solution path */
285 auto path(std::make_shared<PathGeometric>(si_));
286 for (int i = mpath.size() - 1; i >= 0; --i)
287 path->append(mpath[i]->state_);
288 // Add the solution path.
289 base::PlannerSolution psol(path);
290 psol.setPlannerName(getName());
291 if (approximate)
292 psol.setApproximate(approxdif);
293 pdef_->addSolutionPath(psol);
294 solved = true;
295 }
296
297 si_->freeState(xstate);
298 if (rmotion->state_ != nullptr)
299 si_->freeState(rmotion->state_);
300 delete rmotion;
301
302 OMPL_INFORM("%s: Created %u states", getName().c_str(), statesGenerated);
303
304 return {solved, approximate};
305}
306
308{
309 // optimization - check if the bounded approximation invariant
310 // will be violated after the edge insertion (at least for the child node)
311 // if this is the case - perform the local planning
312 // this prevents the update of the graph due to the edge insertion and then the re-update as it is removed
313 double potential_cost = parent->costLb_ + c;
314 if (child->costApx_ > (1 + epsilon_) * potential_cost)
315 if (!checkMotion(parent, child))
316 return;
317
318 // update lowerBoundGraph_
319 std::list<std::size_t> affected;
320
321 lowerBoundGraph_.addEdge(parent->id_, child->id_, c, true, affected);
322
323 // now, check if the bounded apprimation invariant has been violated for each affected vertex
324 // insert them into a priority queue ordered according to the lb cost
325 std::list<std::size_t>::iterator iter;
326 IsLessThanLB isLessThanLB(this);
327 std::set<Motion *, IsLessThanLB> queue(isLessThanLB);
328
329 for (iter = affected.begin(); iter != affected.end(); ++iter)
330 {
331 Motion *m = getMotion(*iter);
332 m->costLb_ = lowerBoundGraph_.getShortestPathCost(*iter);
333 if (m->costApx_ > (1 + epsilon_) * m->costLb_)
334 queue.insert(m);
335 }
336
337 while (!queue.empty())
338 {
339 Motion *motion = *(queue.begin());
340 queue.erase(queue.begin());
341
342 if (motion->costApx_ > (1 + epsilon_) * motion->costLb_)
343 {
344 Motion *potential_parent = getMotion(lowerBoundGraph_.getShortestPathParent(motion->id_));
345 if (checkMotion(potential_parent, motion))
346 {
347 double delta = lazilyUpdateApxParent(motion, potential_parent);
348 updateChildCostsApx(motion, delta);
349 }
350 else
351 {
352 affected.clear();
353
354 lowerBoundGraph_.removeEdge(potential_parent->id_, motion->id_, true, affected);
355
356 for (iter = affected.begin(); iter != affected.end(); ++iter)
357 {
358 Motion *affected = getMotion(*iter);
359 auto lb_queue_iter = queue.find(affected);
360 if (lb_queue_iter != queue.end())
361 {
362 queue.erase(lb_queue_iter);
363 affected->costLb_ = lowerBoundGraph_.getShortestPathCost(affected->id_);
364 if (affected->costApx_ > (1 + epsilon_) * affected->costLb_)
365 queue.insert(affected);
366 }
367 else
368 {
369 affected->costLb_ = lowerBoundGraph_.getShortestPathCost(affected->id_);
370 }
371 }
372
373 motion->costLb_ = lowerBoundGraph_.getShortestPathCost(motion->id_);
374 if (motion->costApx_ > (1 + epsilon_) * motion->costLb_)
375 queue.insert(motion);
376
377 // optimization - we can remove the opposite edge
378 lowerBoundGraph_.removeEdge(motion->id_, potential_parent->id_, false, affected);
379 }
380 }
381 }
382
383 }
384
386{
387 Planner::getPlannerData(data);
388
389 std::vector<Motion *> motions;
390 if (nn_)
391 nn_->list(motions);
392
393 if (lastGoalMotion_ != nullptr)
394 data.addGoalVertex(base::PlannerDataVertex(lastGoalMotion_->state_));
395
396 for (auto &motion : motions)
397 {
398 if (motion->parentApx_ == nullptr)
399 data.addStartVertex(base::PlannerDataVertex(motion->state_));
400 else
401 data.addEdge(base::PlannerDataVertex(motion->parentApx_->state_), base::PlannerDataVertex(motion->state_));
402 }
403}
404
406{
407 for (auto child : m->childrenApx_)
408 {
409 child->costApx_ += delta;
410 updateChildCostsApx(child, delta);
411 }
412}
413
415{
416 double dist = distanceFunction(parent, child);
417 removeFromParentApx(child);
418 double deltaApx = parent->costApx_ + dist - child->costApx_;
419 child->parentApx_ = parent;
420 parent->childrenApx_.push_back(child);
421 child->costApx_ = parent->costApx_ + dist;
422
423 return deltaApx;
424}
425
427{
428 std::vector<Motion *> &vec = m->parentApx_->childrenApx_;
429 for (auto it = vec.begin(); it != vec.end(); ++it)
430 if (*it == m)
431 {
432 vec.erase(it);
433 break;
434 }
435}
Abstract definition of a goal region that can be sampled.
Abstract definition of goals.
Definition: Goal.h:63
virtual bool isSatisfied(const State *st) const =0
Return true if the state satisfies the goal constraints.
Base class for a vertex in the PlannerData structure. All derived classes must implement the clone an...
Definition: PlannerData.h:59
Object containing planner generated vertex and edge data. It is assumed that all vertices are unique,...
Definition: PlannerData.h:175
unsigned int addStartVertex(const PlannerDataVertex &v)
Adds the given vertex to the graph data, and marks it as a start vertex. The vertex index is returned...
unsigned int addGoalVertex(const PlannerDataVertex &v)
Adds the given vertex to the graph data, and marks it as a start vertex. The vertex index is returned...
virtual bool addEdge(unsigned int v1, unsigned int v2, const PlannerDataEdge &edge=PlannerDataEdge(), Cost weight=Cost(1.0))
Adds a directed edge between the given vertex indexes. An optional edge structure and weight can be s...
Encapsulate a termination condition for a motion planner. Planners will call operator() to decide whe...
void addPlannerProgressProperty(const std::string &progressPropertyName, const PlannerProgressProperty &prop)
Add a planner progress property called progressPropertyName with a property querying function prop to...
Definition: Planner.h:410
PlannerSpecs specs_
The specifications of the planner (its capabilities)
Definition: Planner.h:429
Definition of an abstract state.
Definition: State.h:50
Representation of a motion.
Definition: LBTRRT.h:160
std::size_t id_
unique id of the motion
Definition: LBTRRT.h:175
double costLb_
The lower bound cost of the motion while it is stored in the lowerBoundGraph_ and this may seem redun...
Definition: LBTRRT.h:181
base::State * state_
The state contained by the motion.
Definition: LBTRRT.h:173
Motion * parentApx_
The parent motion in the approximation tree.
Definition: LBTRRT.h:183
double costApx_
The approximation cost.
Definition: LBTRRT.h:185
std::vector< Motion * > childrenApx_
The children in the approximation tree.
Definition: LBTRRT.h:187
void setup() override
Perform extra configuration steps, if needed. This call will also issue a call to ompl::base::SpaceIn...
Definition: LBTRRT.cpp:84
double getGoalBias() const
Get the goal bias the planner is using.
Definition: LBTRRT.h:97
void removeFromParentApx(Motion *m)
remove motion from its parent in the approximation tree
Definition: LBTRRT.cpp:426
void setApproximationFactor(double epsilon)
Set the apprimation factor.
Definition: LBTRRT.h:132
double getRange() const
Get the range the planner is using.
Definition: LBTRRT.h:113
double lazilyUpdateApxParent(Motion *child, Motion *parent)
lazily update the parent in the approximation tree without updating costs to cildren
Definition: LBTRRT.cpp:414
void setRange(double distance)
Set the range the planner is supposed to use.
Definition: LBTRRT.h:107
void getPlannerData(base::PlannerData &data) const override
Get information about the current run of the motion planner. Repeated calls to this function will upd...
Definition: LBTRRT.cpp:385
base::PlannerStatus solve(const base::PlannerTerminationCondition &ptc) override
Function that can solve the motion planning problem. This function can be called multiple times on th...
Definition: LBTRRT.cpp:112
void clear() override
Clear all internal datastructures. Planner settings are not affected. Subsequent calls to solve() wil...
Definition: LBTRRT.cpp:70
void freeMemory()
Free the memory allocated by this planner.
Definition: LBTRRT.cpp:98
double getApproximationFactor() const
Get the apprimation factor.
Definition: LBTRRT.h:138
void considerEdge(Motion *parent, Motion *child, double c)
consider an edge for addition to the roadmap
Definition: LBTRRT.cpp:307
LBTRRT(const base::SpaceInformationPtr &si)
Constructor.
Definition: LBTRRT.cpp:44
void setGoalBias(double goalBias)
Set the goal bias.
Definition: LBTRRT.h:91
void updateChildCostsApx(Motion *m, double delta)
update the child cost of the approximation tree
Definition: LBTRRT.cpp:405
This class contains methods that automatically configure various parameters for motion planning....
Definition: SelfConfig.h:60
void configurePlannerRange(double &range)
Compute what a good length for motion segments is.
Definition: SelfConfig.cpp:225
#define OMPL_INFORM(fmt,...)
Log a formatted information string.
Definition: Console.h:68
#define OMPL_ERROR(fmt,...)
Log a formatted error string.
Definition: Console.h:64
Representation of a solution to a planning problem.
void setApproximate(double difference)
Specify that the solution is approximate and set the difference to the goal.
void setPlannerName(const std::string &name)
Set the name of the planner used to compute this solution.
bool directed
Flag indicating whether the planner is able to account for the fact that the validity of a motion fro...
Definition: Planner.h:212
bool approximateSolutions
Flag indicating whether the planner is able to compute approximate solutions.
Definition: Planner.h:202
A class to store the exit status of Planner::solve()
Definition: PlannerStatus.h:49
@ INVALID_START
Invalid start state or no start state specified.
Definition: PlannerStatus.h:56
@ INVALID_GOAL
Invalid goal state.
Definition: PlannerStatus.h:58
comparator - metric is the lower bound cost
Definition: LBTRRT.h:213
comparator - metric is the cost to reach state via a specific state
Definition: LBTRRT.h:192