1 /* 2 * Copyright (C) 2008 The Android Open Source Project 3 * 4 * Licensed under the Apache License, Version 2.0 (the "License"); 5 * you may not use this file except in compliance with the License. 6 * You may obtain a copy of the License at 7 * 8 * http://www.apache.org/licenses/LICENSE-2.0 9 * 10 * Unless required by applicable law or agreed to in writing, software 11 * distributed under the License is distributed on an "AS IS" BASIS, 12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 13 * See the License for the specific language governing permissions and 14 * limitations under the License. 15 */ 16 17 package android.hardware; 18 19 /** 20 * This class represents a {@link android.hardware.Sensor Sensor} event and 21 * holds informations such as the sensor's type, the time-stamp, accuracy and of 22 * course the sensor's {@link SensorEvent#values data}. 23 * 24 * <p> 25 * <u>Definition of the coordinate system used by the SensorEvent API.</u> 26 * </p> 27 * 28 * <p> 29 * The coordinate-system is defined relative to the screen of the phone in its 30 * default orientation. The axes are not swapped when the device's screen 31 * orientation changes. 32 * </p> 33 * 34 * <p> 35 * The X axis is horizontal and points to the right, the Y axis is vertical and 36 * points up and the Z axis points towards the outside of the front face of the 37 * screen. In this system, coordinates behind the screen have negative Z values. 38 * </p> 39 * 40 * <p> 41 * <center><img src="../../../images/axis_device.png" 42 * alt="Sensors coordinate-system diagram." border="0" /></center> 43 * </p> 44 * 45 * <p> 46 * <b>Note:</b> This coordinate system is different from the one used in the 47 * Android 2D APIs where the origin is in the top-left corner. 48 * </p> 49 * 50 * @see SensorManager 51 * @see SensorEvent 52 * @see Sensor 53 * 54 */ 55 56 public class SensorEvent { 57 /** 58 * <p> 59 * The length and contents of the {@link #values values} array depends on 60 * which {@link android.hardware.Sensor sensor} type is being monitored (see 61 * also {@link SensorEvent} for a definition of the coordinate system used). 62 * </p> 63 * 64 * <h4>{@link android.hardware.Sensor#TYPE_ACCELEROMETER 65 * Sensor.TYPE_ACCELEROMETER}:</h4> All values are in SI units (m/s^2) 66 * 67 * <ul> 68 * <li> values[0]: Acceleration minus Gx on the x-axis </li> 69 * <li> values[1]: Acceleration minus Gy on the y-axis </li> 70 * <li> values[2]: Acceleration minus Gz on the z-axis </li> 71 * </ul> 72 * 73 * <p> 74 * A sensor of this type measures the acceleration applied to the device 75 * (<b>Ad</b>). Conceptually, it does so by measuring forces applied to the 76 * sensor itself (<b>Fs</b>) using the relation: 77 * </p> 78 * 79 * <b><center>Ad = - ∑Fs / mass</center></b> 80 * 81 * <p> 82 * In particular, the force of gravity is always influencing the measured 83 * acceleration: 84 * </p> 85 * 86 * <b><center>Ad = -g - ∑F / mass</center></b> 87 * 88 * <p> 89 * For this reason, when the device is sitting on a table (and obviously not 90 * accelerating), the accelerometer reads a magnitude of <b>g</b> = 9.81 91 * m/s^2 92 * </p> 93 * 94 * <p> 95 * Similarly, when the device is in free-fall and therefore dangerously 96 * accelerating towards to ground at 9.81 m/s^2, its accelerometer reads a 97 * magnitude of 0 m/s^2. 98 * </p> 99 * 100 * <p> 101 * It should be apparent that in order to measure the real acceleration of 102 * the device, the contribution of the force of gravity must be eliminated. 103 * This can be achieved by applying a <i>high-pass</i> filter. Conversely, a 104 * <i>low-pass</i> filter can be used to isolate the force of gravity. 105 * </p> 106 * 107 * <pre class="prettyprint"> 108 * 109 * public void onSensorChanged(SensorEvent event) 110 * { 111 * // alpha is calculated as t / (t + dT) 112 * // with t, the low-pass filter's time-constant 113 * // and dT, the event delivery rate 114 * 115 * final float alpha = 0.8; 116 * 117 * gravity[0] = alpha * gravity[0] + (1 - alpha) * event.values[0]; 118 * gravity[1] = alpha * gravity[1] + (1 - alpha) * event.values[1]; 119 * gravity[2] = alpha * gravity[2] + (1 - alpha) * event.values[2]; 120 * 121 * linear_acceleration[0] = event.values[0] - gravity[0]; 122 * linear_acceleration[1] = event.values[1] - gravity[1]; 123 * linear_acceleration[2] = event.values[2] - gravity[2]; 124 * } 125 * </pre> 126 * 127 * <p> 128 * <u>Examples</u>: 129 * <ul> 130 * <li>When the device lies flat on a table and is pushed on its left side 131 * toward the right, the x acceleration value is positive.</li> 132 * 133 * <li>When the device lies flat on a table, the acceleration value is 134 * +9.81, which correspond to the acceleration of the device (0 m/s^2) minus 135 * the force of gravity (-9.81 m/s^2).</li> 136 * 137 * <li>When the device lies flat on a table and is pushed toward the sky 138 * with an acceleration of A m/s^2, the acceleration value is equal to 139 * A+9.81 which correspond to the acceleration of the device (+A m/s^2) 140 * minus the force of gravity (-9.81 m/s^2).</li> 141 * </ul> 142 * 143 * 144 * <h4>{@link android.hardware.Sensor#TYPE_MAGNETIC_FIELD 145 * Sensor.TYPE_MAGNETIC_FIELD}:</h4> 146 * All values are in micro-Tesla (uT) and measure the ambient magnetic field 147 * in the X, Y and Z axis. 148 * 149 * <h4>{@link android.hardware.Sensor#TYPE_GYROSCOPE Sensor.TYPE_GYROSCOPE}: 150 * </h4> All values are in radians/second and measure the rate of rotation 151 * around the device's local X, Y and Z axis. The coordinate system is the 152 * same as is used for the acceleration sensor. Rotation is positive in the 153 * counter-clockwise direction. That is, an observer looking from some 154 * positive location on the x, y or z axis at a device positioned on the 155 * origin would report positive rotation if the device appeared to be 156 * rotating counter clockwise. Note that this is the standard mathematical 157 * definition of positive rotation and does not agree with the definition of 158 * roll given earlier. 159 * <ul> 160 * <li> values[0]: Angular speed around the x-axis </li> 161 * <li> values[1]: Angular speed around the y-axis </li> 162 * <li> values[2]: Angular speed around the z-axis </li> 163 * </ul> 164 * <p> 165 * Typically the output of the gyroscope is integrated over time to 166 * calculate a rotation describing the change of angles over the timestep, 167 * for example: 168 * </p> 169 * 170 * <pre class="prettyprint"> 171 * private static final float NS2S = 1.0f / 1000000000.0f; 172 * private final float[] deltaRotationVector = new float[4](); 173 * private float timestamp; 174 * 175 * public void onSensorChanged(SensorEvent event) { 176 * // This timestep's delta rotation to be multiplied by the current rotation 177 * // after computing it from the gyro sample data. 178 * if (timestamp != 0) { 179 * final float dT = (event.timestamp - timestamp) * NS2S; 180 * // Axis of the rotation sample, not normalized yet. 181 * float axisX = event.values[0]; 182 * float axisY = event.values[1]; 183 * float axisZ = event.values[2]; 184 * 185 * // Calculate the angular speed of the sample 186 * float omegaMagnitude = sqrt(axisX*axisX + axisY*axisY + axisZ*axisZ); 187 * 188 * // Normalize the rotation vector if it's big enough to get the axis 189 * if (omegaMagnitude > EPSILON) { 190 * axisX /= omegaMagnitude; 191 * axisY /= omegaMagnitude; 192 * axisZ /= omegaMagnitude; 193 * } 194 * 195 * // Integrate around this axis with the angular speed by the timestep 196 * // in order to get a delta rotation from this sample over the timestep 197 * // We will convert this axis-angle representation of the delta rotation 198 * // into a quaternion before turning it into the rotation matrix. 199 * float thetaOverTwo = omegaMagnitude * dT / 2.0f; 200 * float sinThetaOverTwo = sin(thetaOverTwo); 201 * float cosThetaOverTwo = cos(thetaOverTwo); 202 * deltaRotationVector[0] = sinThetaOverTwo * axisX; 203 * deltaRotationVector[1] = sinThetaOverTwo * axisY; 204 * deltaRotationVector[2] = sinThetaOverTwo * axisZ; 205 * deltaRotationVector[3] = cosThetaOverTwo; 206 * } 207 * timestamp = event.timestamp; 208 * float[] deltaRotationMatrix = new float[9]; 209 * SensorManager.getRotationMatrixFromVector(deltaRotationMatrix, deltaRotationVector); 210 * // User code should concatenate the delta rotation we computed with the current rotation 211 * // in order to get the updated rotation. 212 * // rotationCurrent = rotationCurrent * deltaRotationMatrix; 213 * } 214 * </pre> 215 * <p> 216 * In practice, the gyroscope noise and offset will introduce some errors 217 * which need to be compensated for. This is usually done using the 218 * information from other sensors, but is beyond the scope of this document. 219 * </p> 220 * <h4>{@link android.hardware.Sensor#TYPE_LIGHT Sensor.TYPE_LIGHT}:</h4> 221 * <ul> 222 * <li>values[0]: Ambient light level in SI lux units </li> 223 * </ul> 224 * 225 * <h4>{@link android.hardware.Sensor#TYPE_PRESSURE Sensor.TYPE_PRESSURE}:</h4> 226 * <ul> 227 * <li>values[0]: Atmospheric pressure in hPa (millibar) </li> 228 * </ul> 229 * 230 * <h4>{@link android.hardware.Sensor#TYPE_PROXIMITY Sensor.TYPE_PROXIMITY}: 231 * </h4> 232 * 233 * <ul> 234 * <li>values[0]: Proximity sensor distance measured in centimeters </li> 235 * </ul> 236 * 237 * <p> 238 * <b>Note:</b> Some proximity sensors only support a binary <i>near</i> or 239 * <i>far</i> measurement. In this case, the sensor should report its 240 * {@link android.hardware.Sensor#getMaximumRange() maximum range} value in 241 * the <i>far</i> state and a lesser value in the <i>near</i> state. 242 * </p> 243 * 244 * <h4>{@link android.hardware.Sensor#TYPE_GRAVITY Sensor.TYPE_GRAVITY}:</h4> 245 * <p>A three dimensional vector indicating the direction and magnitude of gravity. Units 246 * are m/s^2. The coordinate system is the same as is used by the acceleration sensor.</p> 247 * <p><b>Note:</b> When the device is at rest, the output of the gravity sensor should be identical 248 * to that of the accelerometer.</p> 249 * 250 * <h4>{@link android.hardware.Sensor#TYPE_LINEAR_ACCELERATION Sensor.TYPE_LINEAR_ACCELERATION}:</h4> 251 * A three dimensional vector indicating acceleration along each device axis, not including 252 * gravity. All values have units of m/s^2. The coordinate system is the same as is used by the 253 * acceleration sensor. 254 * <p>The output of the accelerometer, gravity and linear-acceleration sensors must obey the 255 * following relation:</p> 256 * <p><ul>acceleration = gravity + linear-acceleration</ul></p> 257 * 258 * <h4>{@link android.hardware.Sensor#TYPE_ROTATION_VECTOR Sensor.TYPE_ROTATION_VECTOR}:</h4> 259 * <p>The rotation vector represents the orientation of the device as a combination of an <i>angle</i> 260 * and an <i>axis</i>, in which the device has rotated through an angle θ around an axis 261 * <x, y, z>.</p> 262 * <p>The three elements of the rotation vector are 263 * <x*sin(θ/2), y*sin(θ/2), z*sin(θ/2)>, such that the magnitude of the rotation 264 * vector is equal to sin(θ/2), and the direction of the rotation vector is equal to the 265 * direction of the axis of rotation.</p> 266 * </p>The three elements of the rotation vector are equal to 267 * the last three components of a <b>unit</b> quaternion 268 * <cos(θ/2), x*sin(θ/2), y*sin(θ/2), z*sin(θ/2)>.</p> 269 * <p>Elements of the rotation vector are unitless. 270 * The x,y, and z axis are defined in the same way as the acceleration 271 * sensor.</p> 272 * The reference coordinate system is defined as a direct orthonormal basis, 273 * where: 274 * </p> 275 * 276 * <ul> 277 * <li>X is defined as the vector product <b>Y.Z</b> (It is tangential to 278 * the ground at the device's current location and roughly points East).</li> 279 * <li>Y is tangential to the ground at the device's current location and 280 * points towards magnetic north.</li> 281 * <li>Z points towards the sky and is perpendicular to the ground.</li> 282 * </ul> 283 * 284 * <p> 285 * <center><img src="../../../images/axis_globe.png" 286 * alt="World coordinate-system diagram." border="0" /></center> 287 * </p> 288 * 289 * <ul> 290 * <li> values[0]: x*sin(θ/2) </li> 291 * <li> values[1]: y*sin(θ/2) </li> 292 * <li> values[2]: z*sin(θ/2) </li> 293 * <li> values[3]: cos(θ/2) </li> 294 * <li> values[4]: estimated heading Accuracy (in radians) (-1 if unavailable)</li> 295 * </ul> 296 * <p> values[3], originally optional, will always be present from SDK Level 18 onwards. 297 * values[4] is a new value that has been added in SDK Level 18. 298 * </p> 299 * 300 * <h4>{@link android.hardware.Sensor#TYPE_ORIENTATION 301 * Sensor.TYPE_ORIENTATION}:</h4> All values are angles in degrees. 302 * 303 * <ul> 304 * <li> values[0]: Azimuth, angle between the magnetic north direction and the 305 * y-axis, around the z-axis (0 to 359). 0=North, 90=East, 180=South, 306 * 270=West 307 * </p> 308 * 309 * <p> 310 * values[1]: Pitch, rotation around x-axis (-180 to 180), with positive 311 * values when the z-axis moves <b>toward</b> the y-axis. 312 * </p> 313 * 314 * <p> 315 * values[2]: Roll, rotation around the y-axis (-90 to 90) 316 * increasing as the device moves clockwise. 317 * </p> 318 * </ul> 319 * 320 * <p> 321 * <b>Note:</b> This definition is different from <b>yaw, pitch and roll</b> 322 * used in aviation where the X axis is along the long side of the plane 323 * (tail to nose). 324 * </p> 325 * 326 * <p> 327 * <b>Note:</b> This sensor type exists for legacy reasons, please use 328 * {@link android.hardware.Sensor#TYPE_ROTATION_VECTOR 329 * rotation vector sensor type} and 330 * {@link android.hardware.SensorManager#getRotationMatrix 331 * getRotationMatrix()} in conjunction with 332 * {@link android.hardware.SensorManager#remapCoordinateSystem 333 * remapCoordinateSystem()} and 334 * {@link android.hardware.SensorManager#getOrientation getOrientation()} to 335 * compute these values instead. 336 * </p> 337 * 338 * <p> 339 * <b>Important note:</b> For historical reasons the roll angle is positive 340 * in the clockwise direction (mathematically speaking, it should be 341 * positive in the counter-clockwise direction). 342 * </p> 343 * 344 * <h4>{@link android.hardware.Sensor#TYPE_RELATIVE_HUMIDITY 345 * Sensor.TYPE_RELATIVE_HUMIDITY}:</h4> 346 * <ul> 347 * <li> values[0]: Relative ambient air humidity in percent </li> 348 * </ul> 349 * <p> 350 * When relative ambient air humidity and ambient temperature are 351 * measured, the dew point and absolute humidity can be calculated. 352 * </p> 353 * <u>Dew Point</u> 354 * <p> 355 * The dew point is the temperature to which a given parcel of air must be 356 * cooled, at constant barometric pressure, for water vapor to condense 357 * into water. 358 * </p> 359 * <center><pre> 360 * ln(RH/100%) + m·t/(T<sub>n</sub>+t) 361 * t<sub>d</sub>(t,RH) = T<sub>n</sub> · ------------------------------ 362 * m - [ln(RH/100%) + m·t/(T<sub>n</sub>+t)] 363 * </pre></center> 364 * <dl> 365 * <dt>t<sub>d</sub></dt> <dd>dew point temperature in °C</dd> 366 * <dt>t</dt> <dd>actual temperature in °C</dd> 367 * <dt>RH</dt> <dd>actual relative humidity in %</dd> 368 * <dt>m</dt> <dd>17.62</dd> 369 * <dt>T<sub>n</sub></dt> <dd>243.12 °C</dd> 370 * </dl> 371 * <p>for example:</p> 372 * <pre class="prettyprint"> 373 * h = Math.log(rh / 100.0) + (17.62 * t) / (243.12 + t); 374 * td = 243.12 * h / (17.62 - h); 375 * </pre> 376 * <u>Absolute Humidity</u> 377 * <p> 378 * The absolute humidity is the mass of water vapor in a particular volume 379 * of dry air. The unit is g/m<sup>3</sup>. 380 * </p> 381 * <center><pre> 382 * RH/100%·A·exp(m·t/(T<sub>n</sub>+t)) 383 * d<sub>v</sub>(t,RH) = 216.7 · ------------------------- 384 * 273.15 + t 385 * </pre></center> 386 * <dl> 387 * <dt>d<sub>v</sub></dt> <dd>absolute humidity in g/m<sup>3</sup></dd> 388 * <dt>t</dt> <dd>actual temperature in °C</dd> 389 * <dt>RH</dt> <dd>actual relative humidity in %</dd> 390 * <dt>m</dt> <dd>17.62</dd> 391 * <dt>T<sub>n</sub></dt> <dd>243.12 °C</dd> 392 * <dt>A</dt> <dd>6.112 hPa</dd> 393 * </dl> 394 * <p>for example:</p> 395 * <pre class="prettyprint"> 396 * dv = 216.7 * 397 * (rh / 100.0 * 6.112 * Math.exp(17.62 * t / (243.12 + t)) / (273.15 + t)); 398 * </pre> 399 * 400 * <h4>{@link android.hardware.Sensor#TYPE_AMBIENT_TEMPERATURE Sensor.TYPE_AMBIENT_TEMPERATURE}: 401 * </h4> 402 * 403 * <ul> 404 * <li> values[0]: ambient (room) temperature in degree Celsius.</li> 405 * </ul> 406 * 407 * 408 * <h4>{@link android.hardware.Sensor#TYPE_MAGNETIC_FIELD_UNCALIBRATED 409 * Sensor.TYPE_MAGNETIC_FIELD_UNCALIBRATED}:</h4> 410 * Similar to {@link android.hardware.Sensor#TYPE_MAGNETIC_FIELD}, 411 * but the hard iron calibration is reported separately instead of being included 412 * in the measurement. Factory calibration and temperature compensation will still 413 * be applied to the "uncalibrated" measurement. Assumptions that the magnetic field 414 * is due to the Earth's poles is avoided. 415 * <p> 416 * The values array is shown below: 417 * <ul> 418 * <li> values[0] = x_uncalib </li> 419 * <li> values[1] = y_uncalib </li> 420 * <li> values[2] = z_uncalib </li> 421 * <li> values[3] = x_bias </li> 422 * <li> values[4] = y_bias </li> 423 * <li> values[5] = z_bias </li> 424 * </ul> 425 * </p> 426 * <p> 427 * x_uncalib, y_uncalib, z_uncalib are the measured magnetic field in X, Y, Z axes. 428 * Soft iron and temperature calibrations are applied. But the hard iron 429 * calibration is not applied. The values are in micro-Tesla (uT). 430 * </p> 431 * <p> 432 * x_bias, y_bias, z_bias give the iron bias estimated in X, Y, Z axes. 433 * Each field is a component of the estimated hard iron calibration. 434 * The values are in micro-Tesla (uT). 435 * </p> 436 * <p> Hard iron - These distortions arise due to the magnetized iron, steel or permanenet 437 * magnets on the device. 438 * Soft iron - These distortions arise due to the interaction with the earth's magentic 439 * field. 440 * </p> 441 * <h4> {@link android.hardware.Sensor#TYPE_GAME_ROTATION_VECTOR}:</h4> 442 * Identical to {@link android.hardware.Sensor#TYPE_ROTATION_VECTOR} except that it 443 * doesn't use the geomagnetic field. Therefore the Y axis doesn't 444 * point north, but instead to some other reference, that reference is 445 * allowed to drift by the same order of magnitude as the gyroscope 446 * drift around the Z axis. 447 * <p> 448 * In the ideal case, a phone rotated and returning to the same real-world 449 * orientation will report the same game rotation vector 450 * (without using the earth's geomagnetic field). However, the orientation 451 * may drift somewhat over time. See {@link android.hardware.Sensor#TYPE_ROTATION_VECTOR} 452 * for a detailed description of the values. This sensor will not have 453 * the estimated heading accuracy value. 454 * </p> 455 * 456 * <h4> {@link android.hardware.Sensor#TYPE_GYROSCOPE_UNCALIBRATED 457 * Sensor.TYPE_GYROSCOPE_UNCALIBRATED}:</h4> 458 * All values are in radians/second and measure the rate of rotation 459 * around the X, Y and Z axis. An estimation of the drift on each axis is 460 * reported as well. 461 * <p> 462 * No gyro-drift compensation is performed. Factory calibration and temperature 463 * compensation is still applied to the rate of rotation (angular speeds). 464 * </p> 465 * <p> 466 * The coordinate system is the same as is used for the 467 * {@link android.hardware.Sensor#TYPE_ACCELEROMETER} 468 * Rotation is positive in the counter-clockwise direction (right-hand rule). 469 * That is, an observer looking from some positive location on the x, y or z axis 470 * at a device positioned on the origin would report positive rotation if the device 471 * appeared to be rotating counter clockwise. 472 * The range would at least be 17.45 rad/s (ie: ~1000 deg/s). 473 * <ul> 474 * <li> values[0] : angular speed (w/o drift compensation) around the X axis in rad/s </li> 475 * <li> values[1] : angular speed (w/o drift compensation) around the Y axis in rad/s </li> 476 * <li> values[2] : angular speed (w/o drift compensation) around the Z axis in rad/s </li> 477 * <li> values[3] : estimated drift around X axis in rad/s </li> 478 * <li> values[4] : estimated drift around Y axis in rad/s </li> 479 * <li> values[5] : estimated drift around Z axis in rad/s </li> 480 * </ul> 481 * </p> 482 * <p><b>Pro Tip:</b> Always use the length of the values array while performing operations 483 * on it. In earlier versions, this used to be always 3 which has changed now. </p> 484 * 485 * @see GeomagneticField 486 */ 487 public final float[] values; 488 489 /** 490 * The sensor that generated this event. See 491 * {@link android.hardware.SensorManager SensorManager} for details. 492 */ 493 public Sensor sensor; 494 495 /** 496 * The accuracy of this event. See {@link android.hardware.SensorManager 497 * SensorManager} for details. 498 */ 499 public int accuracy; 500 501 /** 502 * The time in nanosecond at which the event happened 503 */ 504 public long timestamp; 505 506 SensorEvent(int valueSize) { 507 values = new float[valueSize]; 508 } 509 } 510
