Directory: | ./ |
---|---|
File: | include/crocoddyl/core/state-base.hpp |
Date: | 2025-01-16 08:47:40 |
Exec | Total | Coverage | |
---|---|---|---|
Lines: | 2 | 2 | 100.0% |
Branches: | 5 | 6 | 83.3% |
Line | Branch | Exec | Source |
---|---|---|---|
1 | /////////////////////////////////////////////////////////////////////////////// | ||
2 | // BSD 3-Clause License | ||
3 | // | ||
4 | // Copyright (C) 2019-2020, LAAS-CNRS, University of Edinburgh, INRIA | ||
5 | // Copyright note valid unless otherwise stated in individual files. | ||
6 | // All rights reserved. | ||
7 | /////////////////////////////////////////////////////////////////////////////// | ||
8 | |||
9 | #ifndef CROCODDYL_CORE_STATE_BASE_HPP_ | ||
10 | #define CROCODDYL_CORE_STATE_BASE_HPP_ | ||
11 | |||
12 | #include <stdexcept> | ||
13 | #include <string> | ||
14 | #include <vector> | ||
15 | |||
16 | #include "crocoddyl/core/fwd.hpp" | ||
17 | #include "crocoddyl/core/mathbase.hpp" | ||
18 | #include "crocoddyl/core/utils/exception.hpp" | ||
19 | |||
20 | namespace crocoddyl { | ||
21 | |||
22 | enum Jcomponent { both = 0, first = 1, second = 2 }; | ||
23 | |||
24 | 61761 | inline bool is_a_Jcomponent(Jcomponent firstsecond) { | |
25 |
5/6✓ Branch 0 taken 46339 times.
✓ Branch 1 taken 15422 times.
✓ Branch 2 taken 442 times.
✓ Branch 3 taken 45897 times.
✓ Branch 4 taken 442 times.
✗ Branch 5 not taken.
|
61761 | return (firstsecond == first || firstsecond == second || firstsecond == both); |
26 | } | ||
27 | |||
28 | /** | ||
29 | * @brief Abstract class for the state representation | ||
30 | * | ||
31 | * A state is represented by its operators: difference, integrates, transport | ||
32 | * and their derivatives. The difference operator returns the value of | ||
33 | * \f$\mathbf{x}_{1}\ominus\mathbf{x}_{0}\f$ operation. Instead the integrate | ||
34 | * operator returns the value of \f$\mathbf{x}\oplus\delta\mathbf{x}\f$. These | ||
35 | * operators are used to compared two points on the state manifold | ||
36 | * \f$\mathcal{M}\f$ or to advance the state given a tangential velocity | ||
37 | * (\f$T_\mathbf{x} \mathcal{M}\f$). Therefore the points \f$\mathbf{x}\f$, | ||
38 | * \f$\mathbf{x}_{0}\f$ and \f$\mathbf{x}_{1}\f$ belong to the manifold | ||
39 | * \f$\mathcal{M}\f$; and \f$\delta\mathbf{x}\f$ or | ||
40 | * \f$\mathbf{x}_{1}\ominus\mathbf{x}_{0}\f$ lie on its tangential space. | ||
41 | * | ||
42 | * \sa `diff()`, `integrate()`, `Jdiff()`, `Jintegrate()` and | ||
43 | * `JintegrateTransport()` | ||
44 | */ | ||
45 | template <typename _Scalar> | ||
46 | class StateAbstractTpl { | ||
47 | public: | ||
48 | EIGEN_MAKE_ALIGNED_OPERATOR_NEW | ||
49 | |||
50 | typedef _Scalar Scalar; | ||
51 | typedef MathBaseTpl<Scalar> MathBase; | ||
52 | typedef typename MathBase::VectorXs VectorXs; | ||
53 | typedef typename MathBase::MatrixXs MatrixXs; | ||
54 | |||
55 | /** | ||
56 | * @brief Initialize the state dimensions | ||
57 | * | ||
58 | * @param[in] nx Dimension of state configuration tuple | ||
59 | * @param[in] ndx Dimension of state tangent vector | ||
60 | */ | ||
61 | StateAbstractTpl(const std::size_t nx, const std::size_t ndx); | ||
62 | StateAbstractTpl(); | ||
63 | virtual ~StateAbstractTpl(); | ||
64 | |||
65 | /** | ||
66 | * @brief Generate a zero state | ||
67 | */ | ||
68 | virtual VectorXs zero() const = 0; | ||
69 | |||
70 | /** | ||
71 | * @brief Generate a random state | ||
72 | */ | ||
73 | virtual VectorXs rand() const = 0; | ||
74 | |||
75 | /** | ||
76 | * @brief Compute the state manifold differentiation. | ||
77 | * | ||
78 | * The state differentiation is defined as: | ||
79 | * \f{equation*}{ | ||
80 | * \delta\mathbf{x} = \mathbf{x}_{1} \ominus \mathbf{x}_{0}, | ||
81 | * \f} | ||
82 | * where \f$\mathbf{x}_{1}\f$, \f$\mathbf{x}_{0}\f$ are the current and | ||
83 | * previous state which lie in a manifold \f$\mathcal{M}\f$, and | ||
84 | * \f$\delta\mathbf{x} \in T_\mathbf{x} \mathcal{M}\f$ is the rate of change | ||
85 | * in the state in the tangent space of the manifold. | ||
86 | * | ||
87 | * @param[in] x0 Previous state point (size `nx`) | ||
88 | * @param[in] x1 Current state point (size `nx`) | ||
89 | * @param[out] dxout Difference between the current and previous state points | ||
90 | * (size `ndx`) | ||
91 | */ | ||
92 | virtual void diff(const Eigen::Ref<const VectorXs>& x0, | ||
93 | const Eigen::Ref<const VectorXs>& x1, | ||
94 | Eigen::Ref<VectorXs> dxout) const = 0; | ||
95 | |||
96 | /** | ||
97 | * @brief Compute the state manifold integration. | ||
98 | * | ||
99 | * The state integration is defined as: | ||
100 | * \f{equation*}{ | ||
101 | * \mathbf{x}_{next} = \mathbf{x} \oplus \delta\mathbf{x}, | ||
102 | * \f} | ||
103 | * where \f$\mathbf{x}\f$, \f$\mathbf{x}_{next}\f$ are the current and next | ||
104 | * state which lie in a manifold \f$\mathcal{M}\f$, and \f$\delta\mathbf{x} | ||
105 | * \in T_\mathbf{x} \mathcal{M}\f$ is the rate of change in the state in the | ||
106 | * tangent space of the manifold. | ||
107 | * | ||
108 | * @param[in] x State point (size `nx`) | ||
109 | * @param[in] dx Velocity vector (size `ndx`) | ||
110 | * @param[out] xout Next state point (size `nx`) | ||
111 | */ | ||
112 | virtual void integrate(const Eigen::Ref<const VectorXs>& x, | ||
113 | const Eigen::Ref<const VectorXs>& dx, | ||
114 | Eigen::Ref<VectorXs> xout) const = 0; | ||
115 | |||
116 | /** | ||
117 | * @brief Compute the Jacobian of the state manifold differentiation. | ||
118 | * | ||
119 | * The state differentiation is defined as: | ||
120 | * \f{equation*}{ | ||
121 | * \delta\mathbf{x} = \mathbf{x}_{1} \ominus \mathbf{x}_{0}, | ||
122 | * \f} | ||
123 | * where \f$\mathbf{x}_{1}\f$, \f$\mathbf{x}_{0}\f$ are the current and | ||
124 | * previous state which lie in a manifold \f$\mathcal{M}\f$, and | ||
125 | * \f$\delta\mathbf{x} \in T_\mathbf{x} \mathcal{M}\f$ is the rate of change | ||
126 | * in the state in the tangent space of the manifold. | ||
127 | * | ||
128 | * The Jacobians lie in the tangent space of manifold, i.e. | ||
129 | * \f$\mathbb{R}^{\textrm{ndx}\times\textrm{ndx}}\f$. Note that the state is | ||
130 | * represented as a tuple of `nx` values and its dimension is `ndx`. Calling | ||
131 | * \f$\boldsymbol{\Delta}(\mathbf{x}_{0}, \mathbf{x}_{1}) \f$, the difference | ||
132 | * function, these Jacobians satisfy the following relationships: | ||
133 | * - | ||
134 | * \f$\boldsymbol{\Delta}(\mathbf{x}_{0},\mathbf{x}_{0}\oplus\delta\mathbf{y}) | ||
135 | * - \boldsymbol{\Delta}(\mathbf{x}_{0},\mathbf{x}_{1}) = | ||
136 | * \mathbf{J}_{\mathbf{x}_{1}}\delta\mathbf{y} + | ||
137 | * \mathbf{o}(\mathbf{x}_{0})\f$. | ||
138 | * - | ||
139 | * \f$\boldsymbol{\Delta}(\mathbf{x}_{0}\oplus\delta\mathbf{y},\mathbf{x}_{1}) | ||
140 | * - \boldsymbol{\Delta}(\mathbf{x}_{0},\mathbf{x}_{1}) = | ||
141 | * \mathbf{J}_{\mathbf{x}_{0}}\delta\mathbf{y} + | ||
142 | * \mathbf{o}(\mathbf{x}_{0})\f$, | ||
143 | * | ||
144 | * where \f$\mathbf{J}_{\mathbf{x}_{1}}\f$ and | ||
145 | * \f$\mathbf{J}_{\mathbf{x}_{0}}\f$ are the Jacobian with respect to the | ||
146 | * current and previous state, respectively. | ||
147 | * | ||
148 | * @param[in] x0 Previous state point (size `nx`) | ||
149 | * @param[in] x1 Current state point (size `nx`) | ||
150 | * @param[out] Jfirst Jacobian of the difference operation relative to | ||
151 | * the previous state point (size `ndx`\f$\times\f$`ndx`) | ||
152 | * @param[out] Jsecond Jacobian of the difference operation relative to | ||
153 | * the current state point (size `ndx`\f$\times\f$`ndx`) | ||
154 | * @param[in] firstsecond Argument (either x0 and / or x1) with respect to | ||
155 | * which the differentiation is performed. | ||
156 | */ | ||
157 | virtual void Jdiff(const Eigen::Ref<const VectorXs>& x0, | ||
158 | const Eigen::Ref<const VectorXs>& x1, | ||
159 | Eigen::Ref<MatrixXs> Jfirst, Eigen::Ref<MatrixXs> Jsecond, | ||
160 | const Jcomponent firstsecond = both) const = 0; | ||
161 | |||
162 | /** | ||
163 | * @brief Compute the Jacobian of the state manifold integration. | ||
164 | * | ||
165 | * The state integration is defined as: | ||
166 | * \f{equation*}{ | ||
167 | * \mathbf{x}_{next} = \mathbf{x} \oplus \delta\mathbf{x}, | ||
168 | * \f} | ||
169 | * where \f$\mathbf{x}\f$, \f$\mathbf{x}_{next}\f$ are the current and next | ||
170 | * state which lie in a manifold \f$\mathcal{M}\f$, and \f$\delta\mathbf{x} | ||
171 | * \in T_\mathbf{x} \mathcal{M}\f$ is the rate of change in the state in the | ||
172 | * tangent space of the manifold. | ||
173 | * | ||
174 | * The Jacobians lie in the tangent space of manifold, i.e. | ||
175 | * \f$\mathbb{R}^{\textrm{ndx}\times\textrm{ndx}}\f$. Note that the state is | ||
176 | * represented as a tuple of `nx` values and its dimension is `ndx`. Calling | ||
177 | * \f$ \mathbf{f}(\mathbf{x}, \delta\mathbf{x}) \f$, the integrate function, | ||
178 | * these Jacobians satisfy the following relationships: | ||
179 | * - | ||
180 | * \f$\mathbf{f}(\mathbf{x}\oplus\delta\mathbf{y},\delta\mathbf{x})\ominus\mathbf{f}(\mathbf{x},\delta\mathbf{x}) | ||
181 | * = \mathbf{J}_\mathbf{x}\delta\mathbf{y} + \mathbf{o}(\delta\mathbf{x})\f$. | ||
182 | * - | ||
183 | * \f$\mathbf{f}(\mathbf{x},\delta\mathbf{x}+\delta\mathbf{y})\ominus\mathbf{f}(\mathbf{x},\delta\mathbf{x}) | ||
184 | * = \mathbf{J}_{\delta\mathbf{x}}\delta\mathbf{y} + | ||
185 | * \mathbf{o}(\delta\mathbf{x})\f$, | ||
186 | * | ||
187 | * where \f$\mathbf{J}_{\delta\mathbf{x}}\f$ and \f$\mathbf{J}_{\mathbf{x}}\f$ | ||
188 | * are the Jacobian with respect to the state and velocity, respectively. | ||
189 | * | ||
190 | * @param[in] x State point (size `nx`) | ||
191 | * @param[in] dx Velocity vector (size `ndx`) | ||
192 | * @param[out] Jfirst Jacobian of the integration operation relative to | ||
193 | * the state point (size `ndx`\f$\times\f$`ndx`) | ||
194 | * @param[out] Jsecond Jacobian of the integration operation relative to | ||
195 | * the velocity vector (size `ndx`\f$\times\f$`ndx`) | ||
196 | * @param[in] firstsecond Argument (either x and / or dx) with respect to | ||
197 | * which the differentiation is performed | ||
198 | * @param[in] op Assignment operator which sets, adds, or removes | ||
199 | * the given Jacobian matrix | ||
200 | */ | ||
201 | virtual void Jintegrate(const Eigen::Ref<const VectorXs>& x, | ||
202 | const Eigen::Ref<const VectorXs>& dx, | ||
203 | Eigen::Ref<MatrixXs> Jfirst, | ||
204 | Eigen::Ref<MatrixXs> Jsecond, | ||
205 | const Jcomponent firstsecond = both, | ||
206 | const AssignmentOp op = setto) const = 0; | ||
207 | |||
208 | /** | ||
209 | * @brief Parallel transport from integrate(x, dx) to x. | ||
210 | * | ||
211 | * This function performs the parallel transportation of an input matrix whose | ||
212 | * columns are expressed in the tangent space at | ||
213 | * \f$\mathbf{x}\oplus\delta\mathbf{x}\f$ to the tangent space at | ||
214 | * \f$\mathbf{x}\f$ point. | ||
215 | * | ||
216 | * @param[in] x State point (size `nx`). | ||
217 | * @param[in] dx Velocity vector (size `ndx`) | ||
218 | * @param[out] Jin Input matrix (number of rows = `nv`). | ||
219 | * @param[in] firstsecond Argument (either x or dx) with respect to which the | ||
220 | * differentiation of Jintegrate is performed. | ||
221 | */ | ||
222 | virtual void JintegrateTransport(const Eigen::Ref<const VectorXs>& x, | ||
223 | const Eigen::Ref<const VectorXs>& dx, | ||
224 | Eigen::Ref<MatrixXs> Jin, | ||
225 | const Jcomponent firstsecond) const = 0; | ||
226 | |||
227 | /** | ||
228 | * @copybrief diff() | ||
229 | * | ||
230 | * @param[in] x0 Previous state point (size `nx`) | ||
231 | * @param[in] x1 Current state point (size `nx`) | ||
232 | * @return Difference between the current and previous state points (size | ||
233 | * `ndx`) | ||
234 | */ | ||
235 | VectorXs diff_dx(const Eigen::Ref<const VectorXs>& x0, | ||
236 | const Eigen::Ref<const VectorXs>& x1); | ||
237 | |||
238 | /** | ||
239 | * @copybrief integrate() | ||
240 | * | ||
241 | * @param[in] x State point (size `nx`) | ||
242 | * @param[in] dx Velocity vector (size `ndx`) | ||
243 | * @return Next state point (size `nx`) | ||
244 | */ | ||
245 | VectorXs integrate_x(const Eigen::Ref<const VectorXs>& x, | ||
246 | const Eigen::Ref<const VectorXs>& dx); | ||
247 | |||
248 | /** | ||
249 | * @copybrief jdiff() | ||
250 | * | ||
251 | * @param[in] x0 Previous state point (size `nx`) | ||
252 | * @param[in] x1 Current state point (size `nx`) | ||
253 | * @return Jacobians | ||
254 | */ | ||
255 | std::vector<MatrixXs> Jdiff_Js(const Eigen::Ref<const VectorXs>& x0, | ||
256 | const Eigen::Ref<const VectorXs>& x1, | ||
257 | const Jcomponent firstsecond = both); | ||
258 | |||
259 | /** | ||
260 | * @copybrief Jintegrate() | ||
261 | * | ||
262 | * @param[in] x State point (size `nx`) | ||
263 | * @param[in] dx Velocity vector (size `ndx`) | ||
264 | * @return Jacobians | ||
265 | */ | ||
266 | std::vector<MatrixXs> Jintegrate_Js(const Eigen::Ref<const VectorXs>& x, | ||
267 | const Eigen::Ref<const VectorXs>& dx, | ||
268 | const Jcomponent firstsecond = both); | ||
269 | |||
270 | /** | ||
271 | * @brief Return the dimension of the state tuple | ||
272 | */ | ||
273 | std::size_t get_nx() const; | ||
274 | |||
275 | /** | ||
276 | * @brief Return the dimension of the tangent space of the state manifold | ||
277 | */ | ||
278 | std::size_t get_ndx() const; | ||
279 | |||
280 | /** | ||
281 | * @brief Return the dimension of the configuration tuple | ||
282 | */ | ||
283 | std::size_t get_nq() const; | ||
284 | |||
285 | /** | ||
286 | * @brief Return the dimension of tangent space of the configuration manifold | ||
287 | */ | ||
288 | std::size_t get_nv() const; | ||
289 | |||
290 | /** | ||
291 | * @brief Return the state lower bound | ||
292 | */ | ||
293 | const VectorXs& get_lb() const; | ||
294 | |||
295 | /** | ||
296 | * @brief Return the state upper bound | ||
297 | */ | ||
298 | const VectorXs& get_ub() const; | ||
299 | |||
300 | /** | ||
301 | * @brief Indicate if the state has defined limits | ||
302 | */ | ||
303 | bool get_has_limits() const; | ||
304 | |||
305 | /** | ||
306 | * @brief Modify the state lower bound | ||
307 | */ | ||
308 | void set_lb(const VectorXs& lb); | ||
309 | |||
310 | /** | ||
311 | * @brief Modify the state upper bound | ||
312 | */ | ||
313 | void set_ub(const VectorXs& ub); | ||
314 | |||
315 | protected: | ||
316 | void update_has_limits(); | ||
317 | |||
318 | std::size_t nx_; //!< State dimension | ||
319 | std::size_t ndx_; //!< State rate dimension | ||
320 | std::size_t nq_; //!< Configuration dimension | ||
321 | std::size_t nv_; //!< Velocity dimension | ||
322 | VectorXs lb_; //!< Lower state limits | ||
323 | VectorXs ub_; //!< Upper state limits | ||
324 | bool has_limits_; //!< Indicates whether any of the state limits is finite | ||
325 | }; | ||
326 | |||
327 | } // namespace crocoddyl | ||
328 | |||
329 | /* --- Details -------------------------------------------------------------- */ | ||
330 | /* --- Details -------------------------------------------------------------- */ | ||
331 | /* --- Details -------------------------------------------------------------- */ | ||
332 | #include "crocoddyl/core/state-base.hxx" | ||
333 | |||
334 | #endif // CROCODDYL_CORE_STATE_BASE_HPP_ | ||
335 |