px4_autopilot/EKF/terrain_estimator.cpp

/****************************************************************************
 *
 *   Copyright (c) 2015 Estimation and Control Library (ECL). 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 ECL 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 terrain_estimator.cpp
 * Function for fusing rangefinder measurements to estimate terrain vertical position/
 *
 * @author Paul Riseborough <p_riseborough@live.com.au>
 *
 */

#include "ekf.h"
#include <ecl.h>
#include <mathlib/mathlib.h>

bool Ekf::initHagl()
{
	bool initialized = false;

	if (!_control_status.flags.in_air) {
		// if on ground, do not trust the range sensor, but assume a ground clearance
		_terrain_vpos = _state.pos(2) + _params.rng_gnd_clearance;
		// use the ground clearance value as our uncertainty
		_terrain_var = sq(_params.rng_gnd_clearance);
		_time_last_fake_hagl_fuse = _time_last_imu;
		initialized = true;

	} else if (shouldUseRangeFinderForHagl()
		   && _range_sensor.isDataHealthy()) {
		// if we have a fresh measurement, use it to initialise the terrain estimator
		_terrain_vpos = _state.pos(2) + _range_sensor.getDistBottom();
		// initialise state variance to variance of measurement
		_terrain_var = sq(_params.range_noise);
		// success
		initialized = true;

	} else if (shouldUseOpticalFlowForHagl()
		   && _flow_for_terrain_data_ready) {
		// initialise terrain vertical position to origin as this is the best guess we have
		_terrain_vpos = fmaxf(0.0f,  _state.pos(2));
		_terrain_var = 100.0f;
		initialized = true;

	} else {
		// no information - cannot initialise
	}

	if (initialized) {
		// has initialized with valid data
		_time_last_hagl_fuse = _time_last_imu;
	}

	return initialized;
}

void Ekf::runTerrainEstimator()
{
	// If we are on ground, store the local position and time to use as a reference
	if (!_control_status.flags.in_air) {
		_last_on_ground_posD = _state.pos(2);
	}

	// Perform initialisation check and
	// on ground, continuously reset the terrain estimator
	if (!_terrain_initialised || !_control_status.flags.in_air) {
		_terrain_initialised = initHagl();

	} else {

		// predict the state variance growth where the state is the vertical position of the terrain underneath the vehicle

		// process noise due to errors in vehicle height estimate
		_terrain_var += sq(_imu_sample_delayed.delta_vel_dt * _params.terrain_p_noise);

		// process noise due to terrain gradient
		_terrain_var += sq(_imu_sample_delayed.delta_vel_dt * _params.terrain_gradient)
				* (sq(_state.vel(0)) + sq(_state.vel(1)));

		// limit the variance to prevent it becoming badly conditioned
		_terrain_var = math::constrain(_terrain_var, 0.0f, 1e4f);

		// Fuse range finder data if available
		if (shouldUseRangeFinderForHagl()
		    && _range_sensor.isDataHealthy()) {
			fuseHagl();
		}

		if (shouldUseOpticalFlowForHagl()
		    && _flow_for_terrain_data_ready) {
			fuseFlowForTerrain();
			_flow_for_terrain_data_ready = false;
		}

		// constrain _terrain_vpos to be a minimum of _params.rng_gnd_clearance larger than _state.pos(2)
		if (_terrain_vpos - _state.pos(2) < _params.rng_gnd_clearance) {
			_terrain_vpos = _params.rng_gnd_clearance + _state.pos(2);
		}
	}

	updateTerrainValidity();
}

void Ekf::fuseHagl()
{
	// get a height above ground measurement from the range finder assuming a flat earth
	const float meas_hagl = _range_sensor.getDistBottom();

	// predict the hagl from the vehicle position and terrain height
	const float pred_hagl = _terrain_vpos - _state.pos(2);

	// calculate the innovation
	_hagl_innov = pred_hagl - meas_hagl;

	// calculate the observation variance adding the variance of the vehicles own height uncertainty
	const float obs_variance = fmaxf(P(9,9) * _params.vehicle_variance_scaler, 0.0f)
			     + sq(_params.range_noise)
			     + sq(_params.range_noise_scaler * _range_sensor.getRange());

	// calculate the innovation variance - limiting it to prevent a badly conditioned fusion
	_hagl_innov_var = fmaxf(_terrain_var + obs_variance, obs_variance);

	// perform an innovation consistency check and only fuse data if it passes
	const float gate_size = fmaxf(_params.range_innov_gate, 1.0f);
	_hagl_test_ratio = sq(_hagl_innov) / (sq(gate_size) * _hagl_innov_var);

	if (_hagl_test_ratio <= 1.0f) {
		// calculate the Kalman gain
		const float gain = _terrain_var / _hagl_innov_var;
		// correct the state
		_terrain_vpos -= gain * _hagl_innov;
		// correct the variance
		_terrain_var = fmaxf(_terrain_var * (1.0f - gain), 0.0f);
		// record last successful fusion event
		_time_last_hagl_fuse = _time_last_imu;
		_innov_check_fail_status.flags.reject_hagl = false;

	} else {
		// If we have been rejecting range data for too long, reset to measurement
		const uint64_t timeout = static_cast<uint64_t>(_params.terrain_timeout * 1e6f);
		if (isTimedOut(_time_last_hagl_fuse, timeout)) {
			_terrain_vpos = _state.pos(2) + meas_hagl;
			_terrain_var = obs_variance;
			_terrain_vpos_reset_counter++;

		} else {
			_innov_check_fail_status.flags.reject_hagl = true;
		}
	}
}

void Ekf::fuseFlowForTerrain()
{
	// calculate optical LOS rates using optical flow rates that have had the body angular rate contribution removed
	// correct for gyro bias errors in the data used to do the motion compensation
	// Note the sign convention used: A positive LOS rate is a RH rotation of the scene about that axis.
	const Vector2f opt_flow_rate = _flow_compensated_XY_rad / _flow_sample_delayed.dt + Vector2f(_flow_gyro_bias);

	// get latest estimated orientation
	const float q0 = _state.quat_nominal(0);
	const float q1 = _state.quat_nominal(1);
	const float q2 = _state.quat_nominal(2);
	const float q3 = _state.quat_nominal(3);

	// calculate the optical flow observation variance
	const float R_LOS = calcOptFlowMeasVar();

	// get rotation matrix from earth to body
	const Dcmf earth_to_body = quatToInverseRotMat(_state.quat_nominal);

	// calculate the sensor position relative to the IMU
	const Vector3f pos_offset_body = _params.flow_pos_body - _params.imu_pos_body;

	// calculate the velocity of the sensor relative to the imu in body frame
	// Note: _flow_sample_delayed.gyro_xyz is the negative of the body angular velocity, thus use minus sign
	const Vector3f vel_rel_imu_body = Vector3f(-_flow_sample_delayed.gyro_xyz / _flow_sample_delayed.dt) % pos_offset_body;

	// calculate the velocity of the sensor in the earth frame
	const Vector3f vel_rel_earth = _state.vel + _R_to_earth * vel_rel_imu_body;

	// rotate into body frame
	const Vector3f vel_body = earth_to_body * vel_rel_earth;

	const float t0 = q0 * q0 - q1 * q1 - q2 * q2 + q3 * q3;

	// constrain terrain to minimum allowed value and predict height above ground
	_terrain_vpos = fmaxf(_terrain_vpos, _params.rng_gnd_clearance + _state.pos(2));
	const float pred_hagl_inv = 1.f / (_terrain_vpos - _state.pos(2));

	// Calculate observation matrix for flow around the vehicle x axis
	const float Hx = vel_body(1) * t0 * pred_hagl_inv * pred_hagl_inv;

	// Constrain terrain variance to be non-negative
	_terrain_var = fmaxf(_terrain_var, 0.0f);

	// Cacluate innovation variance
	_flow_innov_var(0) = Hx * Hx * _terrain_var + R_LOS;

	// calculate the kalman gain for the flow x measurement
	const float Kx = _terrain_var * Hx / _flow_innov_var(0);

	// calculate prediced optical flow about x axis
	const float pred_flow_x = vel_body(1) * earth_to_body(2, 2) * pred_hagl_inv;

	// calculate flow innovation (x axis)
	_flow_innov(0) = pred_flow_x - opt_flow_rate(0);

	// calculate correction term for terrain variance
	const float KxHxP =  Kx * Hx * _terrain_var;

	// innovation consistency check
	const float gate_size = fmaxf(_params.flow_innov_gate, 1.0f);
	float flow_test_ratio = sq(_flow_innov(0)) / (sq(gate_size) * _flow_innov_var(0));

	// do not perform measurement update if badly conditioned
	if (flow_test_ratio <= 1.0f) {
		_terrain_vpos += Kx * _flow_innov(0);
		// guard against negative variance
		_terrain_var = fmaxf(_terrain_var - KxHxP, 0.0f);
		_time_last_flow_terrain_fuse = _time_last_imu;
	}

	// Calculate observation matrix for flow around the vehicle y axis
	const float Hy = -vel_body(0) * t0 * pred_hagl_inv * pred_hagl_inv;

	// Calculuate innovation variance
	_flow_innov_var(1) = Hy * Hy * _terrain_var + R_LOS;

	// calculate the kalman gain for the flow y measurement
	const float Ky = _terrain_var * Hy / _flow_innov_var(1);

	// calculate prediced optical flow about y axis
	const float pred_flow_y = -vel_body(0) * earth_to_body(2, 2) * pred_hagl_inv;

	// calculate flow innovation (y axis)
	_flow_innov(1) = pred_flow_y - opt_flow_rate(1);

	// calculate correction term for terrain variance
	const float KyHyP =  Ky * Hy * _terrain_var;

	// innovation consistency check
	flow_test_ratio = sq(_flow_innov(1)) / (sq(gate_size) * _flow_innov_var(1));

	if (flow_test_ratio <= 1.0f) {
		_terrain_vpos += Ky * _flow_innov(1);
		// guard against negative variance
		_terrain_var = fmaxf(_terrain_var - KyHyP, 0.0f);
		_time_last_flow_terrain_fuse = _time_last_imu;
	}
}

void Ekf::updateTerrainValidity()
{
	// we have been fusing range finder measurements in the last 5 seconds
	const bool recent_range_fusion = isRecent(_time_last_hagl_fuse, (uint64_t)5e6);

	// we have been fusing optical flow measurements for terrain estimation within the last 5 seconds
	// this can only be the case if the main filter does not fuse optical flow
	const bool recent_flow_for_terrain_fusion = isRecent(_time_last_flow_terrain_fuse, (uint64_t)5e6);

	_hagl_valid = (_terrain_initialised && (recent_range_fusion || recent_flow_for_terrain_fusion));

	_hagl_sensor_status.flags.range_finder = shouldUseRangeFinderForHagl()
						 && recent_range_fusion
						 && (_time_last_fake_hagl_fuse != _time_last_hagl_fuse);

	_hagl_sensor_status.flags.flow = shouldUseOpticalFlowForHagl() && recent_flow_for_terrain_fusion;
}