test_mathlib.cpp 12.3 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406
/****************************************************************************
 *
 *  Copyright (C) 2013-2019 PX4 Development Team. All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 *
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. 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.
 * 3. Neither the name PX4 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.
 *
 ****************************************************************************/

/**
 * @file test_mathlib.cpp
 * Tests for the PX4 math library.
 */

#include <unit_test.h>

#include <errno.h>
#include <fcntl.h>
#include <float.h>
#include <math.h>
#include <px4_platform_common/px4_config.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <sys/types.h>
#include <unistd.h>

#include <px4_platform_common/log.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <string.h>
#include <time.h>
#include <mathlib/mathlib.h>
#include <systemlib/err.h>
#include <drivers/drv_hrt.h>
#include <matrix/math.hpp>

#include "tests_main.h"

class MathlibTest : public UnitTest
{
public:
	virtual bool run_tests();

private:
	bool testVector2();
	bool testVector3();
	bool testVector10();
	bool testMatrix3x3();
	bool testMatrix10x10();
	bool testMatrixNonsymmetric();
	bool testRotationMatrixQuaternion();
	bool testQuaternionfrom_dcm();
	bool testQuaternionfrom_euler();
	bool testQuaternionRotate();
	bool testFinite();
};

#define TEST_OP(_title, _op) { unsigned int n = 30000; hrt_abstime t0, t1; t0 = hrt_absolute_time(); for (unsigned int j = 0; j < n; j++) { _op; }; t1 = hrt_absolute_time(); PX4_INFO(_title ": %.6fus", (double)(t1 - t0) / n); }

using namespace math;

bool MathlibTest::testVector2()
{
	{
		matrix::Vector2f v;
		matrix::Vector2f v1(1.0f, 2.0f);
		matrix::Vector2f v2(1.0f, -1.0f);
		float data[2] = {1.0f, 2.0f};
		TEST_OP("Constructor matrix::Vector2f()", matrix::Vector2f v3);
		TEST_OP("Constructor matrix::Vector2f(matrix::Vector2f)", matrix::Vector2f v3(v1); ut_assert_true(v3 == v1); v3.zero());
		TEST_OP("Constructor matrix::Vector2f(float[])", matrix::Vector2f v3(data));
		TEST_OP("Constructor matrix::Vector2f(float, float)", matrix::Vector2f v3(1.0f, 2.0f));
		TEST_OP("matrix::Vector2f = matrix::Vector2f", v = v1);
		TEST_OP("matrix::Vector2f + matrix::Vector2f", v + v1);
		TEST_OP("matrix::Vector2f - matrix::Vector2f", v - v1);
		TEST_OP("matrix::Vector2f += matrix::Vector2f", v += v1);
		TEST_OP("matrix::Vector2f -= matrix::Vector2f", v -= v1);
		TEST_OP("matrix::Vector2f * matrix::Vector2f", v * v1);
		TEST_OP("matrix::Vector2f %% matrix::Vector2f", v1 % v2);
	}
	return true;
}

bool MathlibTest::testVector3()
{

	{
		matrix::Vector3f v;
		matrix::Vector3f v1(1.0f, 2.0f, 0.0f);
		matrix::Vector3f v2(1.0f, -1.0f, 2.0f);
		float data[3] = {1.0f, 2.0f, 3.0f};
		TEST_OP("Constructor matrix::Vector3f()", matrix::Vector3f v3);
		TEST_OP("Constructor matrix::Vector3f(matrix::Vector3f)", matrix::Vector3f v3(v1); ut_assert_true(v3 == v1); v3.zero());
		TEST_OP("Constructor matrix::Vector3f(float[])", matrix::Vector3f v3(data));
		TEST_OP("Constructor matrix::Vector3f(float, float, float)", matrix::Vector3f v3(1.0f, 2.0f, 3.0f));
		TEST_OP("matrix::Vector3f = matrix::Vector3f", v = v1);
		TEST_OP("matrix::Vector3f + matrix::Vector3f", v + v1);
		TEST_OP("matrix::Vector3f - matrix::Vector3f", v - v1);
		TEST_OP("matrix::Vector3f += matrix::Vector3f", v += v1);
		TEST_OP("matrix::Vector3f -= matrix::Vector3f", v -= v1);
		TEST_OP("matrix::Vector3f * float", v1 * 2.0f);
		TEST_OP("matrix::Vector3f / float", v1 / 2.0f);
		TEST_OP("matrix::Vector3f *= float", v1 *= 2.0f);
		TEST_OP("matrix::Vector3f /= float", v1 /= 2.0f);
		TEST_OP("matrix::Vector3f * matrix::Vector3f", v * v1);
		TEST_OP("matrix::Vector3f %% matrix::Vector3f", v1 % v2);
		TEST_OP("matrix::Vector3f length", v1.length());
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wunused-variable"
		// Need pragma here instead of moving variable out of TEST_OP and just reference because
		// TEST_OP measures performance of vector operations.
		TEST_OP("matrix::Vector3f element read", volatile float a = v1(0));
#pragma GCC diagnostic pop
		TEST_OP("matrix::Vector3f element write", v1(0) = 1.0f);
	}
	return true;
}

bool MathlibTest::testMatrix3x3()
{
	{
		matrix::Matrix3f m1;
		m1.identity();
		matrix::Matrix3f m2;
		m2.identity();
		matrix::Vector3f v1(1.0f, 2.0f, 0.0f);
		TEST_OP("matrix::Matrix3f * matrix::Vector3f", m1 * v1);
		TEST_OP("matrix::Matrix3f + matrix::Matrix3f", m1 + m2);
		TEST_OP("matrix::Matrix3f * matrix::Matrix3f", m1 * m2);
	}
	return true;
}

bool MathlibTest::testMatrixNonsymmetric()
{
	int rc = true;
	{
		//PX4_INFO("Nonsymmetric matrix operations test");
		// test nonsymmetric +, -, +=, -=

		float data1[2][3] = {{1, 2, 3}, {4, 5, 6}};
		float data2[2][3] = {{2, 4, 6}, {8, 10, 12}};
		float data3[2][3] = {{3, 6, 9}, {12, 15, 18}};

		matrix::Matrix<float, 2, 3> m1(data1);
		matrix::Matrix<float, 2, 3> m2(data2);
		matrix::Matrix<float, 2, 3> m3(data3);

		if (m1 + m2 != m3) {
			PX4_ERR("matrix::Matrix<float, 2, 3> + matrix::Matrix<float, 2, 3> failed!");
			(m1 + m2).print();
			printf("!=\n");
			m3.print();
			rc = false;
		}

		ut_assert("m1 + m2 == m3", m1 + m2 == m3);

		if (m3 - m2 != m1) {
			PX4_ERR("matrix::Matrix<float, 2, 3> - matrix::Matrix<float, 2, 3> failed!");
			(m3 - m2).print();
			printf("!=\n");
			m1.print();
			rc = false;
		}

		ut_assert("m3 - m2 == m1", m3 - m2 == m1);

		m1 += m2;

		if (m1 != m3) {
			PX4_ERR("matrix::Matrix<float, 2, 3> += matrix::Matrix<float, 2, 3> failed!");
			m1.print();
			printf("!=\n");
			m3.print();
			rc = false;
		}

		ut_assert("m1 == m3", m1 == m3);

		m1 -= m2;
		matrix::Matrix<float, 2, 3> m1_orig(data1);

		if (m1 != m1_orig) {
			PX4_ERR("matrix::Matrix<float, 2, 3> -= matrix::Matrix<float, 2, 3> failed!");
			m1.print();
			printf("!=\n");
			m1_orig.print();
			rc = false;
		}

		ut_assert("m1 == m1_orig", m1 == m1_orig);

	}
	return rc;
}

bool MathlibTest::testRotationMatrixQuaternion()
{
	// test conversion rotation matrix to quaternion and back
	matrix::Dcmf R_orig;
	matrix::Dcmf R;
	matrix::Quatf q;
	float diff = 0.2f;
	float tol = 0.00001f;

	//PX4_INFO("Quaternion transformation methods test.");

	for (float roll = -M_PI_F; roll <= M_PI_F; roll += diff) {
		for (float pitch = -M_PI_2_F; pitch <= M_PI_2_F; pitch += diff) {
			for (float yaw = -M_PI_F; yaw <= M_PI_F; yaw += diff) {
				R_orig = matrix::Eulerf(roll, pitch, yaw);
				q = R_orig;
				R = q;

				for (int i = 0; i < 3; i++) {
					for (int j = 0; j < 3; j++) {
						ut_assert("matrix::Quatf method 'from_dcm' or 'to_dcm' outside tolerance!", fabsf(R_orig(i, j) - R(i, j)) < tol);
					}
				}
			}
		}
	}

	return true;
}


bool MathlibTest::testQuaternionfrom_dcm()
{
	// test against some known values
	float tol = 0.0001f;
	matrix::Quatf q_true = {1.0f, 0.0f, 0.0f, 0.0f};

	matrix::Matrix3f R_orig;
	R_orig.identity();

	matrix::Quatf q(R_orig);

	for (unsigned i = 0; i < 4; i++) {
		ut_assert("matrix::Quatf method 'from_dcm()' outside tolerance!", fabsf(q(i) - q_true(i)) < tol);
	}

	return true;
}

bool MathlibTest::testQuaternionfrom_euler()
{
	float tol = 0.0001f;
	matrix::Quatf q_true = {1.0f, 0.0f, 0.0f, 0.0f};

	matrix::Matrix3f R_orig;
	R_orig.identity();

	matrix::Quatf q(R_orig);

	q_true = matrix::Eulerf(0.3f, 0.2f, 0.1f);
	q = {0.9833f, 0.1436f, 0.1060f, 0.0343f};

	for (unsigned i = 0; i < 4; i++) {
		ut_assert("matrix::Quatf method 'from_euler()' outside tolerance!", fabsf(q(i) - q_true(i)) < tol);
	}

	q_true = matrix::Eulerf(-1.5f, -0.2f, 0.5f);
	q = {0.7222f, -0.6391f, -0.2386f, 0.1142f};

	for (unsigned i = 0; i < 4; i++) {
		ut_assert("matrix::Quatf method 'from_euler()' outside tolerance!", fabsf(q(i) - q_true(i)) < tol);
	}

	q_true = matrix::Eulerf(M_PI_2_F, -M_PI_2_F, -M_PI_F / 3);
	q = {0.6830f, 0.1830f, -0.6830f, 0.1830f};

	for (unsigned i = 0; i < 4; i++) {
		ut_assert("matrix::Quatf method 'from_euler()' outside tolerance!", fabsf(q(i) - q_true(i)) < tol);
	}

	return true;
}

bool MathlibTest::testQuaternionRotate()
{
	// test quaternion method "rotate" (rotate vector by quaternion)
	matrix::Vector3f vector = {1.0f, 1.0f, 1.0f};
	matrix::Vector3f vector_q;
	matrix::Vector3f vector_r;
	matrix::Quatf q;
	matrix::Dcmf R;
	float diff = 0.2f;
	float tol = 0.00001f;

	//PX4_INFO("matrix::Quatf vector rotation method test.");

	for (float roll = -M_PI_F; roll <= M_PI_F; roll += diff) {
		for (float pitch = -M_PI_2_F; pitch <= M_PI_2_F; pitch += diff) {
			for (float yaw = -M_PI_F; yaw <= M_PI_F; yaw += diff) {
				R = matrix::Eulerf(roll, pitch, yaw);
				q = matrix::Eulerf(roll, pitch, yaw);
				vector_r = R * vector;
				vector_q = q.conjugate(vector);

				for (int i = 0; i < 3; i++) {
					ut_assert("matrix::Quatf method 'rotate' outside tolerance", fabsf(vector_r(i) - vector_q(i)) < tol);
				}
			}
		}
	}


	// test some values calculated with matlab
	tol = 0.0001f;
	q = matrix::Eulerf(M_PI_2_F, 0.0f, 0.0f);
	vector_q = q.conjugate(vector);
	matrix::Vector3f vector_true = {1.00f, -1.00f, 1.00f};

	for (unsigned i = 0; i < 3; i++) {
		ut_assert("matrix::Quatf method 'rotate' outside tolerance", fabsf(vector_true(i) - vector_q(i)) < tol);
	}

	q = matrix::Eulerf(0.3f, 0.2f, 0.1f);
	vector_q = q.conjugate(vector);
	vector_true = {1.1566, 0.7792, 1.0273};

	for (unsigned i = 0; i < 3; i++) {
		ut_assert("matrix::Quatf method 'rotate' outside tolerance", fabsf(vector_true(i) - vector_q(i)) < tol);
	}

	q = matrix::Eulerf(-1.5f, -0.2f, 0.5f);
	vector_q = q.conjugate(vector);
	vector_true = {0.5095, 1.4956, -0.7096};

	for (unsigned i = 0; i < 3; i++) {
		ut_assert("matrix::Quatf method 'rotate' outside tolerance", fabsf(vector_true(i) - vector_q(i)) < tol);
	}

	q = matrix::Eulerf(M_PI_2_F, -M_PI_2_F, -M_PI_F / 3.0f);
	vector_q = q.conjugate(vector);
	vector_true = { -1.3660, 0.3660, 1.0000};

	for (unsigned i = 0; i < 3; i++) {
		ut_assert("matrix::Quatf method 'rotate' outside tolerance", fabsf(vector_true(i) - vector_q(i)) < tol);
	}

	return true;
}

bool MathlibTest::testFinite()
{
	ut_assert("PX4_ISFINITE(0.0f)", PX4_ISFINITE(0.0f) == true);
	ut_assert("PX4_ISFINITE(-0.0f)", PX4_ISFINITE(-0.0f) == true);
	ut_assert("PX4_ISFINITE(1.0f)", PX4_ISFINITE(1.0f) == true);
	ut_assert("PX4_ISFINITE(-1.0f)", PX4_ISFINITE(-1.0f) == true);

	ut_assert("PX4_ISFINITE(NAN)", PX4_ISFINITE(NAN) == false);
	ut_assert("PX4_ISFINITE(1/0)", PX4_ISFINITE(1.0f / 0.0f) == false);
	ut_assert("PX4_ISFINITE(0/0)", PX4_ISFINITE(0.0f / 0.0f) == false);
	ut_assert("PX4_ISFINITE(INFINITY)", PX4_ISFINITE(INFINITY) == false);
	ut_assert("PX4_ISFINITE(NAN * INFINITY)", PX4_ISFINITE(NAN * INFINITY) == false);
	ut_assert("PX4_ISFINITE(NAN * 1.0f)", PX4_ISFINITE(NAN * 1.0f) == false);
	ut_assert("PX4_ISFINITE(INFINITY * 2.0f)", PX4_ISFINITE(INFINITY * 2.0f) == false);

	return true;
}

bool MathlibTest::run_tests()
{
	ut_run_test(testVector2);
	ut_run_test(testVector3);
	ut_run_test(testMatrix3x3);
	ut_run_test(testMatrixNonsymmetric);
	ut_run_test(testRotationMatrixQuaternion);
	ut_run_test(testQuaternionfrom_dcm);
	ut_run_test(testQuaternionfrom_euler);
	ut_run_test(testQuaternionRotate);
	ut_run_test(testFinite);

	return (_tests_failed == 0);
}

ut_declare_test_c(test_mathlib, MathlibTest)