RigidBodyPlanningWithIK.cpp
1 /*********************************************************************
2 * Software License Agreement (BSD License)
3 *
4 * Copyright (c) 2010, Rice 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 Rice 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: Ioan Sucan */
36 
37 #include <ompl/base/spaces/SE3StateSpace.h>
38 #include <ompl/geometric/SimpleSetup.h>
39 #include <ompl/base/goals/GoalLazySamples.h>
40 #include <ompl/geometric/GeneticSearch.h>
41 
42 #include <ompl/config.h>
43 #include <iostream>
44 
45 namespace ob = ompl::base;
46 namespace og = ompl::geometric;
47 
49 // describe an arbitrary representation of a goal region in SE(3)
50 class MyGoalRegion : public ob::GoalRegion
51 {
52 public:
53 
54  MyGoalRegion(const ob::SpaceInformationPtr &si) : ob::GoalRegion(si)
55  {
56  setThreshold(1e-2);
57  }
58 
59  double distanceGoal(const ob::State *state) const override
60  {
61  // goal region is given by states where x + y = z and orientation is close to identity
62  double d = fabs(state->as<ob::SE3StateSpace::StateType>()->getX()
63  + state->as<ob::SE3StateSpace::StateType>()->getY()
64  - state->as<ob::SE3StateSpace::StateType>()->getZ())
65  + fabs(state->as<ob::SE3StateSpace::StateType>()->rotation().w - 1.0);
66  return d;
67  }
68 
69 };
70 
71 // Goal regions such as the one above cannot be sampled, so
72 // bi-directional trees cannot be used for solving. However, we can
73 // transform such goal regions into ones that can be sampled. The
74 // caveat is that it should be possible to find states in this region
75 // with some other algorithm. Genetic algorithms that essentially
76 // perform inverse kinematics in the general sense can be used:
77 
78 bool regionSamplingWithGS(const ob::SpaceInformationPtr &si, const ob::ProblemDefinitionPtr &pd, const ob::GoalRegion *region, const ob::GoalLazySamples *gls, ob::State *result)
79 {
80  og::GeneticSearch g(si);
81 
82  // we can use a larger time duration for solve(), but we want to demo the ability
83  // of GeneticSearch to continue from where it left off
84  bool cont = false;
85  for (int i = 0 ; i < 100 ; ++i)
86  if (g.solve(0.05, *region, result))
87  {
88  cont = true;
89  break;
90  }
91 
92  if (cont)
93  {
94  std::cout << "Found goal state: " << std::endl;
95  si->printState(result);
96  }
97 
98  // we continue sampling while we are able to find solutions, we have found not more than 2 previous solutions and we have not yet solved the problem
99  return cont && gls->maxSampleCount() < 3 && !pd->hasSolution();
100 }
101 
102 void planWithIK()
103 {
104  // construct the state space we are planning in
105  auto space(std::make_shared<ob::SE3StateSpace>());
106 
107  // set the bounds for the R^3 part of SE(3)
108  ob::RealVectorBounds bounds(3);
109  bounds.setLow(-1);
110  bounds.setHigh(1);
111 
112  space->setBounds(bounds);
113 
114  // define a simple setup class
115  og::SimpleSetup ss(space);
116 
117  // create a random start state
119  start->setXYZ(0, 0, 0);
120  start->rotation().setIdentity();
121  ss.addStartState(start);
122 
123  // define our goal region
124  MyGoalRegion region(ss.getSpaceInformation());
125 
126  // bind a sampling function that fills its argument with a sampled state
127  // and returns true while it can produce new samples we don't need to
128  // check if new samples are different from ones previously computed as
129  // this is pefromed automatically by GoalLazySamples
130  ob::GoalSamplingFn samplingFunction = [&ss, &region](const ob::GoalLazySamples *gls, ob::State *result)
131  {
132  return regionSamplingWithGS(ss.getSpaceInformation(), ss.getProblemDefinition(), &region,
133  gls, result);
134  };
135 
136  // create an instance of GoalLazySamples:
137  auto goal(std::make_shared<ob::GoalLazySamples>(ss.getSpaceInformation(), samplingFunction));
138 
139  // we set a goal that is sampleable, but it in fact corresponds to a region that is not sampleable by default
140  ss.setGoal(goal);
141 
142  // attempt to solve the problem
143  ob::PlannerStatus solved = ss.solve(3.0);
144 
145  if (solved)
146  {
147  std::cout << "Found solution:" << std::endl;
148  // print the path to screen
149  ss.simplifySolution();
150  ss.getSolutionPath().print(std::cout);
151  }
152  else
153  std::cout << "No solution found" << std::endl;
154 
155  // the region variable will now go out of scope. To make sure it is not used in the sampling function any more
156  // (i.e., the sampling thread was able to terminate), we make sure sampling has terminated
157  goal->as<ob::GoalLazySamples>()->stopSampling();
158 }
160 
161 int main(int /*argc*/, char ** /*argv*/)
162 {
163  std::cout << "OMPL version: " << OMPL_VERSION << std::endl;
164 
165  planWithIK();
166 
167  return 0;
168 }
Definition of a goal region.
Definition: GoalRegion.h:111
This namespace contains code that is specific to planning under geometric constraints.
Definition: GeneticSearch.h:79
Definition of a goal region that can be sampled, but the sampling process can be slow....
Definition of an abstract state.
Definition: State.h:113
This namespace contains sampling based planning routines shared by both planning under geometric cons...
Create the set of classes typically needed to solve a geometric problem.
Definition: SimpleSetup.h:126
const T * as() const
Cast this instance to a desired type.
Definition: State.h:162
A class to store the exit status of Planner::solve()
virtual double distanceGoal(const State *st) const =0
Compute the distance to the goal (heuristic). This function is the one used in computing the distance...
ompl::base::CompoundState StateType
Define the type of state allocated by this state space.
Definition: StateSpace.h:641
unsigned int maxSampleCount() const override
Return the maximum number of samples that can be asked for before repeating.
std::function< bool(const GoalLazySamples *, State *)> GoalSamplingFn
Goal sampling function. Returns false when no further calls should be made to it. Fills its second ar...
Definition of a scoped state.
Definition: ScopedState.h:120
Genetic Algorithm for searching valid states.
The lower and upper bounds for an Rn space.
void setThreshold(double threshold)
Set the distance to the goal that is allowed for a state to be considered in the goal region.
Definition: GoalRegion.h:171