12345qiupeng
/
gtsam
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fixed GPS Kitti example, VO works but bad results

release/4.3a0
Luca Carlone 2013-08-12 20:45:44 +00:00
parent 7a027be7e5
commit a518dae06a
5 changed files with 382 additions and 440 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

237
matlab/gtsam_examples/IMUKittiExample.m
View File

@ -1,237 +0,0 @@
%close all
%clc
import gtsam.*;
%% Read metadata and compute relative sensor pose transforms
IMU_metadata = importdata('KittiEquivBiasedImu_metadata.txt');
IMU_metadata = cell2struct(num2cell(IMU_metadata.data), IMU_metadata.colheaders, 2);
IMUinBody = Pose3.Expmap([IMU_metadata.BodyPtx; IMU_metadata.BodyPty; IMU_metadata.BodyPtz;
IMU_metadata.BodyPrx; IMU_metadata.BodyPry; IMU_metadata.BodyPrz; ]);
if ~IMUinBody.equals(Pose3, 1e-5)
error 'Currently only support IMUinBody is identity, i.e. IMU and body frame are the same';
end
VO_metadata = importdata('KittiRelativePose_metadata.txt');
VO_metadata = cell2struct(num2cell(VO_metadata.data), VO_metadata.colheaders, 2);
VOinBody = Pose3.Expmap([VO_metadata.BodyPtx; VO_metadata.BodyPty; VO_metadata.BodyPtz;
VO_metadata.BodyPrx; VO_metadata.BodyPry; VO_metadata.BodyPrz; ]);
GPS_metadata = importdata('KittiGps_metadata.txt');
GPS_metadata = cell2struct(num2cell(GPS_metadata.data), GPS_metadata.colheaders, 2);
GPSinBody = Pose3.Expmap([GPS_metadata.BodyPtx; GPS_metadata.BodyPty; GPS_metadata.BodyPtz;
GPS_metadata.BodyPrx; GPS_metadata.BodyPry; GPS_metadata.BodyPrz; ]);
VOinIMU = IMUinBody.inverse().compose(VOinBody);
GPSinIMU = IMUinBody.inverse().compose(GPSinBody);
%% Read data and change coordinate frame of GPS and VO measurements to IMU frame
IMU_data = importdata('KittiEquivBiasedImu.txt');
IMU_data = cell2struct(num2cell(IMU_data.data), IMU_data.colheaders, 2);
imum = cellfun(@(x) x', num2cell([ [IMU_data.accelX]' [IMU_data.accelY]' [IMU_data.accelZ]' [IMU_data.omegaX]' [IMU_data.omegaY]' [IMU_data.omegaZ]' ], 2), 'UniformOutput', false);
[IMU_data.acc_omega] = deal(imum{:});
IMU_data = rmfield(IMU_data, { 'accelX' 'accelY' 'accelZ' 'omegaX' 'omegaY' 'omegaZ' });
clear imum
VO_data = importdata('KittiRelativePose.txt');
VO_data = cell2struct(num2cell(VO_data.data), VO_data.colheaders, 2);
% Merge relative pose fields and convert to Pose3
logposes = [ [VO_data.dtx]' [VO_data.dty]' [VO_data.dtz]' [VO_data.drx]' [VO_data.dry]' [VO_data.drz]' ];
logposes = num2cell(logposes, 2);
relposes = arrayfun(@(x) {gtsam.Pose3.Expmap(x{:}')}, logposes);
relposes = arrayfun(@(x) {VOinIMU.compose(x{:}).compose(VOinIMU.inverse())}, relposes);
[VO_data.RelativePose] = deal(relposes{:});
VO_data = rmfield(VO_data, { 'dtx' 'dty' 'dtz' 'drx' 'dry' 'drz' });
clear logposes relposes
GPS_data = importdata('KittiGps.txt');
GPS_data = cell2struct(num2cell(GPS_data.data), GPS_data.colheaders, 2);
%% Set initial conditions for the estimated trajectory
disp('TODO: we have GPS so this initialization is not right')
currentPoseGlobal = Pose3; % initial pose is the reference frame (navigation frame)
currentVelocityGlobal = [0;0;0]; % the vehicle is stationary at the beginning
bias_acc = [0;0;0]; % we initialize accelerometer biases to zero
bias_omega = [0;0;0]; % we initialize gyro biases to zero
%% Solver object
isamParams = ISAM2Params;
isamParams.setRelinearizeSkip(1);
isam = gtsam.ISAM2(isamParams);
%% create nonlinear factor graph
factors = NonlinearFactorGraph;
values = Values;
%% Create prior on initial pose, velocity, and biases
sigma_init_x = 1.0;
sigma_init_v = 1.0;
sigma_init_b = 1.0;
values.insert(symbol('x',0), currentPoseGlobal);
values.insert(symbol('v',0), LieVector(currentVelocityGlobal) );
values.insert(symbol('b',0), imuBias.ConstantBias(bias_acc,bias_omega) );
disp('TODO: we have GPS so this initialization is not right')
% Prior on initial pose
factors.add(PriorFactorPose3(symbol('x',0), ...
currentPoseGlobal, noiseModel.Isotropic.Sigma(6, sigma_init_x)));
% Prior on initial velocity
factors.add(PriorFactorLieVector(symbol('v',0), ...
LieVector(currentVelocityGlobal), noiseModel.Isotropic.Sigma(3, sigma_init_v)));
% Prior on initial bias
factors.add(PriorFactorConstantBias(symbol('b',0), ...
imuBias.ConstantBias(bias_acc,bias_omega), noiseModel.Isotropic.Sigma(6, sigma_init_b)));
currentSummarizedMeasurement = [];
% Measurement types:
% 1: VO
% 2: GPS
% 3: IMU
timestamps = sortrows( [ ...
[VO_data.Time]' 1*ones(length([VO_data.Time]), 1); ...
%[GPS_data.Time]' 2*ones(length([GPS_data.Time]), 1); ...
%[IMU_data.Time]' 3*ones(length([IMU_data.Time]), 1); ...
], 1); % this are the time-stamps at which we want to initialize a new node in the graph
%% Main loop:
% (1) we read the measurements
% (2) we create the corresponding factors in the graph
% (3) we solve the graph to obtain and optimal estimate of robot trajectory
% t_previous = 0;%LC
% poseIndex = 0;%LC
% currentPose = Pose3; %LC
position= []; %LC
for measurementIndex = 1:size(timestamps,1)
measurementIndex
% currentPose = currentPose.compose(VO_data(measurementIndex).RelativePose);%LC
%
% position(measurementIndex,:) = currentPose.translation.vector;%LC
% At each non=IMU measurement we initialize a new node in the graph
currentPoseKey = symbol('x',poseIndex);
currentVelKey = symbol('v',poseIndex);
currentBiasKey = symbol('b',poseIndex);
t = timestamps(measurementIndex, 1);
type = timestamps(measurementIndex, 2);
% if type == 3
% % Integrate IMU
%
% if isempty(currentSummarizedMeasurement)
% % Create initial empty summarized measurement
% % we assume that each row of the IMU.txt file has the following structure:
% % timestamp delta_t acc_x acc_y acc_z omega_x omega_y omega_z
% currentBias = isam.calculateEstimate(currentBiasKey - 1);
% currentSummarizedMeasurement = gtsam.ImuFactorPreintegratedMeasurements( ...
% currentBias, IMU_metadata.AccelerometerSigma.^2 * eye(3), ...
% IMU_metadata.GyroscopeSigma.^2 * eye(3), IMU_metadata.IntegrationSigma.^2 * eye(3));
% end
%
% % Accumulate preintegrated measurement
% deltaT = IMU_data(index).dt;
% accMeas = IMU_data(index).acc_omega(1:3);
% omegaMeas = IMU_data(index).acc_omega(4:6);
% currentSummarizedMeasurement.integrateMeasurement(accMeas, omegaMeas, deltaT);
%
% else
% % Create IMU factor
% factors.add(ImuFactor( ...
% currentPoseKey-1, currentVelKey-1, ...
% currentPoseKey, currentVelKey, ...
% currentBiasKey-1, currentSummarizedMeasurement, g, cor_v, ...
% currentSummarizedMeasurement.PreintMeasCov));
%
% % Reset summarized measurement
% currentSummarizedMeasurement = [];
%
% if type == 1
% % Create VO factor
% elseif type == 2
% % Create GPS factor
% end
%
% poseIndex = poseIndex + 1;
% end
% =======================================================================
%% add factor corresponding to GPS measurements (if available at the current time)
% % =======================================================================
% if isempty( find(GPS_data(:,1) == t ) ) == 0 % it is a GPS measurement
% if length( find(GPS_data(:,1)) ) > 1
% error('more GPS measurements at the same time stamp: it should be an error')
% end
%
% index = find(GPS_data(:,1) == t ); % the row of the IMU_data matrix that we have to integrate
% GPSmeas = GPS_data(index,2:4);
%
% noiseModelGPS = ???; % noiseModelGPS.Isotropic.Sigma(6, sigma_init_x))
%
% % add factor
% disp('TODO: is the GPS noise right?')
% factors.add(PriorFactor???(currentPoseKey, GPSmeas, noiseModelGPS) );
% end
% =======================================================================
% %% add factor corresponding to VO measurements (if available at the current time)
% % =======================================================================
% if isempty( find([VO_data.Time] == t, 1) )== 0 % it is a GPS measurement
% if length( find([VO_data.Time] == t) ) > 1
% error('more VO measurements at the same time stamp: it should be an error')
% end
%
% index = find([VO_data.Time] == t, 1); % the row of the IMU_data matrix that we have to integrate
%
% VOpose = VO_data(index).RelativePose;
% noiseModelVO = noiseModel.Diagonal.Sigmas([ IMU_metadata.RotationSigma * [1;1;1]; IMU_metadata.TranslationSigma * [1;1;1] ]);
%
% % add factor
% disp('TODO: is the VO noise right?')
% factors.add(BetweenFactorPose3(lastVOPoseKey, currentPoseKey, VOpose, noiseModelVO));
%
% lastVOPoseKey = currentPoseKey;
% end
% % =======================================================================
%
% disp('TODO: add values')
% % values.insert(, initialPose);
% % values.insert(symbol('v',lastIndex+1), initialVel);
%
% %% Update solver
% % =======================================================================
% isam.update(factors, values);
% factors = NonlinearFactorGraph;
% values = Values;
%
% isam.calculateEstimate(currentPoseKey);
% % M = isam.marginalCovariance(key_pose);
% % =======================================================================
%
% previousPoseKey = currentPoseKey;
% previousVelKey = currentVelKey;
% t_previous = t;
end
figure
plot(position(:,1),position(:,2))
% figure(1)
% hold on;
% plot(positions(1,:), positions(2,:), '-b');
% plot3DTrajectory(isam.calculateEstimate, 'g-');
% axis equal;
% legend('true trajectory', 'traj integrated in body', 'traj integrated in nav')

191
matlab/gtsam_examples/IMUKittiExampleAdvanced.m Normal file
View File

@ -0,0 +1,191 @@
close all
clc
import gtsam.*;
disp('Example of application of ISAM2 for visual-inertial navigation on the KITTI VISION BENCHMARK SUITE (http://www.computervisiononline.com/dataset/kitti-vision-benchmark-suite)')
%% Read metadata and compute relative sensor pose transforms
% IMU metadata
disp('-- Reading sensor metadata')
IMU_metadata = importdata('KittiEquivBiasedImu_metadata.txt');
IMU_metadata = cell2struct(num2cell(IMU_metadata.data), IMU_metadata.colheaders, 2);
IMUinBody = Pose3.Expmap([IMU_metadata.BodyPtx; IMU_metadata.BodyPty; IMU_metadata.BodyPtz;
IMU_metadata.BodyPrx; IMU_metadata.BodyPry; IMU_metadata.BodyPrz; ]);
if ~IMUinBody.equals(Pose3, 1e-5)
error 'Currently only support IMUinBody is identity, i.e. IMU and body frame are the same';
end
% VO metadata
VO_metadata = importdata('KittiRelativePose_metadata.txt');
VO_metadata = cell2struct(num2cell(VO_metadata.data), VO_metadata.colheaders, 2);
VOinBody = Pose3.Expmap([VO_metadata.BodyPtx; VO_metadata.BodyPty; VO_metadata.BodyPtz;
VO_metadata.BodyPrx; VO_metadata.BodyPry; VO_metadata.BodyPrz; ]);
VOinIMU = IMUinBody.inverse().compose(VOinBody);
% GPS metadata
GPS_metadata = importdata('KittiGps_metadata.txt');
GPS_metadata = cell2struct(num2cell(GPS_metadata.data), GPS_metadata.colheaders, 2);
GPSinBody = Pose3.Expmap([GPS_metadata.BodyPtx; GPS_metadata.BodyPty; GPS_metadata.BodyPtz;
GPS_metadata.BodyPrx; GPS_metadata.BodyPry; GPS_metadata.BodyPrz; ]);
GPSinIMU = IMUinBody.inverse().compose(GPSinBody);
%% Read data and change coordinate frame of GPS and VO measurements to IMU frame
disp('-- Reading sensor data from file')
% IMU data
IMU_data = importdata('KittiEquivBiasedImu.txt');
IMU_data = cell2struct(num2cell(IMU_data.data), IMU_data.colheaders, 2);
imum = cellfun(@(x) x', num2cell([ [IMU_data.accelX]' [IMU_data.accelY]' [IMU_data.accelZ]' [IMU_data.omegaX]' [IMU_data.omegaY]' [IMU_data.omegaZ]' ], 2), 'UniformOutput', false);
[IMU_data.acc_omega] = deal(imum{:});
%IMU_data = rmfield(IMU_data, { 'accelX' 'accelY' 'accelZ' 'omegaX' 'omegaY' 'omegaZ' });
clear imum
% VO data
VO_data = importdata('KittiRelativePose.txt');
VO_data = cell2struct(num2cell(VO_data.data), VO_data.colheaders, 2);
% Merge relative pose fields and convert to Pose3
logposes = [ [VO_data.dtx]' [VO_data.dty]' [VO_data.dtz]' [VO_data.drx]' [VO_data.dry]' [VO_data.drz]' ];
logposes = num2cell(logposes, 2);
relposes = arrayfun(@(x) {gtsam.Pose3.Expmap(x{:}')}, logposes);
relposes = arrayfun(@(x) {VOinIMU.compose(x{:}).compose(VOinIMU.inverse())}, relposes);
[VO_data.RelativePose] = deal(relposes{:});
VO_data = rmfield(VO_data, { 'dtx' 'dty' 'dtz' 'drx' 'dry' 'drz' });
noiseModelVO = noiseModel.Diagonal.Sigmas([ VO_metadata.RotationSigma * [1;1;1]; VO_metadata.TranslationSigma * [1;1;1] ]);
clear logposes relposes
% % % GPS data
% % GPS_data = importdata('KittiGps.txt');
% % GPS_data = cell2struct(num2cell(GPS_data.data), GPS_data.colheaders, 2);
% % % Convert GPS from lat/long to meters
% % [ x, y, ~ ] = deg2utm( [GPS_data.Latitude], [GPS_data.Longitude] );
% % for i = 1:numel(x)
% % GPS_data(i).Position = gtsam.Point3(x(i), y(i), GPS_data(i).Altitude);
% % end
% % % % Calculate GPS sigma in meters
% % % [ xSig, ySig, ~ ] = deg2utm( [GPS_data.Latitude] + [GPS_data.PositionSigma], ...
% % % [GPS_data.Longitude] + [GPS_data.PositionSigma]);
% % % xSig = xSig - x;
% % % ySig = ySig - y;
% % %% Start at time of first GPS measurement
% % % firstGPSPose = 1;
%% Get initial conditions for the estimated trajectory
% % % currentPoseGlobal = Pose3(Rot3, GPS_data(firstGPSPose).Position); % initial pose is the reference frame (navigation frame)
currentPoseGlobal = Pose3;
currentVelocityGlobal = LieVector([0;0;0]); % the vehicle is stationary at the beginning
currentBias = imuBias.ConstantBias(zeros(3,1), zeros(3,1));
sigma_init_x = noiseModel.Isotropic.Sigma(6, 0.01);
sigma_init_v = noiseModel.Isotropic.Sigma(3, 1000.0);
sigma_init_b = noiseModel.Isotropic.Sigma(6, 0.01);
g = [0;0;-9.8];
w_coriolis = [0;0;0];
%% Solver object
isamParams = ISAM2Params;
isamParams.setFactorization('QR');
isamParams.setRelinearizeSkip(1);
isam = gtsam.ISAM2(isamParams);
newFactors = NonlinearFactorGraph;
newValues = Values;
%% Main loop:
% (1) we read the measurements
% (2) we create the corresponding factors in the graph
% (3) we solve the graph to obtain and optimal estimate of robot trajectory
timestamps = sortrows( [ ...
[VO_data.Time]' 1*ones(length([VO_data.Time]), 1); ...
% % %[GPS_data.Time]' 2*ones(length([GPS_data.Time]), 1); ...
], 1); % this are the time-stamps at which we want to initialize a new node in the graph
timestamps = timestamps(15:end,:); % there seem to be issues with the initial IMU measurements
IMUtimes = [IMU_data.Time];
VOPoseKeys = []; % here we store the keys of the poses involved in VO (between) factors
for measurementIndex = 1:length(timestamps)
% At each non=IMU measurement we initialize a new node in the graph
currentPoseKey = symbol('x',measurementIndex);
currentVelKey = symbol('v',measurementIndex);
currentBiasKey = symbol('b',measurementIndex);
t = timestamps(measurementIndex, 1);
type = timestamps(measurementIndex, 2);
%% bookkeeping
if type == 1 % we store the keys corresponding to VO measurements
VOPoseKeys = [VOPoseKeys; currentPoseKey];
end
if measurementIndex == 1
%% Create initial estimate and prior on initial pose, velocity, and biases
newValues.insert(currentPoseKey, currentPoseGlobal);
newValues.insert(currentVelKey, currentVelocityGlobal);
newValues.insert(currentBiasKey, currentBias);
newFactors.add(PriorFactorPose3(currentPoseKey, currentPoseGlobal, sigma_init_x));
newFactors.add(PriorFactorLieVector(currentVelKey, currentVelocityGlobal, sigma_init_v));
newFactors.add(PriorFactorConstantBias(currentBiasKey, currentBias, sigma_init_b));
else
t_previous = timestamps(measurementIndex-1, 1);
%% Summarize IMU data between the previous GPS measurement and now
IMUindices = find(IMUtimes >= t_previous & IMUtimes <= t);
if ~isempty(IMUindices) % if there are IMU measurements to integrate
currentSummarizedMeasurement = gtsam.ImuFactorPreintegratedMeasurements( ...
currentBias, IMU_metadata.AccelerometerSigma.^2 * eye(3), ...
IMU_metadata.GyroscopeSigma.^2 * eye(3), IMU_metadata.IntegrationSigma.^2 * eye(3));
for imuIndex = IMUindices
accMeas = [ IMU_data(imuIndex).accelX; IMU_data(imuIndex).accelY; IMU_data(imuIndex).accelZ ];
omegaMeas = [ IMU_data(imuIndex).omegaX; IMU_data(imuIndex).omegaY; IMU_data(imuIndex).omegaZ ];
deltaT = IMU_data(imuIndex).dt;
currentSummarizedMeasurement.integrateMeasurement(accMeas, omegaMeas, deltaT);
end
% Create IMU factor
newFactors.add(ImuFactor( ...
currentPoseKey-1, currentVelKey-1, ...
currentPoseKey, currentVelKey, ...
currentBiasKey, currentSummarizedMeasurement, g, w_coriolis));
else % if there are no IMU measurements
error('no IMU measurements in [t_previous, t]')
end
% LC: sigma_init_b is wrong: this should be some uncertainty on bias evolution given in the IMU metadata
newFactors.add(BetweenFactorConstantBias(currentBiasKey-1, currentBiasKey, imuBias.ConstantBias(zeros(3,1), zeros(3,1)), sigma_init_b));
%% Create GPS factor
if type == 2
newFactors.add(PriorFactorPose3(currentPoseKey, Pose3(currentPoseGlobal.rotation, GPS_data(measurementIndex).Position), ...
noiseModel.Diagonal.Precisions([ zeros(3,1); 1./(GPS_data(measurementIndex).PositionSigma).^2*ones(3,1) ])));
end
%% Create VO factor
if type == 1
VOpose = VO_data(measurementIndex).RelativePose;
newFactors.add(BetweenFactorPose3(VOPoseKeys(end-1), VOPoseKeys(end), VOpose, noiseModelVO));
end
% Add initial value
% newValues.insert(currentPoseKey, Pose3(currentPoseGlobal.rotation, GPS_data(measurementIndex).Position));
newValues.insert(currentPoseKey,currentPoseGlobal);
newValues.insert(currentVelKey, currentVelocityGlobal);
newValues.insert(currentBiasKey, currentBias);
% Update solver
% =======================================================================
isam.update(newFactors, newValues);
newFactors = NonlinearFactorGraph;
newValues = Values;
if rem(measurementIndex,100)==0 % plot every 100 time steps
cla;
plot3DTrajectory(isam.calculateEstimate, 'g-');
axis equal
drawnow;
end
% =======================================================================
currentPoseGlobal = isam.calculateEstimate(currentPoseKey);
currentVelocityGlobal = isam.calculateEstimate(currentVelKey);
currentBias = isam.calculateEstimate(currentBiasKey);
end
end % end main loop

149
matlab/gtsam_examples/IMUKittiExampleGPS.m Normal file
View File

@ -0,0 +1,149 @@
close all
clc
import gtsam.*;
disp('Example of application of ISAM2 for GPS-aided navigation on the KITTI VISION BENCHMARK SUITE (http://www.computervisiononline.com/dataset/kitti-vision-benchmark-suite)')
%% Read metadata and compute relative sensor pose transforms
% IMU metadata
disp('-- Reading sensor metadata')
IMU_metadata = importdata('KittiEquivBiasedImu_metadata.txt');
IMU_metadata = cell2struct(num2cell(IMU_metadata.data), IMU_metadata.colheaders, 2);
IMUinBody = Pose3.Expmap([IMU_metadata.BodyPtx; IMU_metadata.BodyPty; IMU_metadata.BodyPtz;
IMU_metadata.BodyPrx; IMU_metadata.BodyPry; IMU_metadata.BodyPrz; ]);
if ~IMUinBody.equals(Pose3, 1e-5)
error 'Currently only support IMUinBody is identity, i.e. IMU and body frame are the same';
end
% GPS metadata
GPS_metadata = importdata('KittiRelativePose_metadata.txt');
GPS_metadata = cell2struct(num2cell(GPS_metadata.data), GPS_metadata.colheaders, 2);
%% Read data
disp('-- Reading sensor data from file')
% IMU data
IMU_data = importdata('KittiEquivBiasedImu.txt');
IMU_data = cell2struct(num2cell(IMU_data.data), IMU_data.colheaders, 2);
imum = cellfun(@(x) x', num2cell([ [IMU_data.accelX]' [IMU_data.accelY]' [IMU_data.accelZ]' [IMU_data.omegaX]' [IMU_data.omegaY]' [IMU_data.omegaZ]' ], 2), 'UniformOutput', false);
[IMU_data.acc_omega] = deal(imum{:});
clear imum
% GPS data
GPS_data = importdata('Gps_converted.txt');
GPS_data = cell2struct(num2cell(GPS_data.data), GPS_data.colheaders, 2);
for i = 1:numel(GPS_data)
GPS_data(i).Position = gtsam.Point3(GPS_data(i).X, GPS_data(i).Y, GPS_data(i).Z);
end
noiseModelGPS = noiseModel.Diagonal.Precisions([ [0;0;0]; 1.0/0.07 * [1;1;1] ]);
firstGPSPose = 2;
GPSskip = 10; % Skip this many GPS measurements each time
%% Get initial conditions for the estimated trajectory
currentPoseGlobal = Pose3(Rot3, GPS_data(firstGPSPose).Position); % initial pose is the reference frame (navigation frame)
currentVelocityGlobal = LieVector([0;0;0]); % the vehicle is stationary at the beginning
currentBias = imuBias.ConstantBias(zeros(3,1), zeros(3,1));
sigma_init_x = noiseModel.Isotropic.Precisions([ 0.0; 0.0; 0.0; 1; 1; 1 ]);
sigma_init_v = noiseModel.Isotropic.Sigma(3, 1000.0);
sigma_init_b = noiseModel.Isotropic.Sigmas([ 0.100; 0.100; 0.100; 5.00e-05; 5.00e-05; 5.00e-05 ]);
sigma_between_b = [ IMU_metadata.AccelerometerBiasSigma * ones(3,1); IMU_metadata.GyroscopeBiasSigma * ones(3,1) ];
g = [0;0;-9.8];
w_coriolis = [0;0;0];
%% Solver object
isamParams = ISAM2Params;
isamParams.setFactorization('CHOLESKY');
isamParams.setRelinearizeSkip(10);
isam = gtsam.ISAM2(isamParams);
newFactors = NonlinearFactorGraph;
newValues = Values;
%% Main loop:
% (1) we read the measurements
% (2) we create the corresponding factors in the graph
% (3) we solve the graph to obtain and optimal estimate of robot trajectory
IMUtimes = [IMU_data.Time];
disp('-- Starting main loop: inference is performed at each time step, but we plot trajectory every 10 steps')
for measurementIndex = firstGPSPose:length(GPS_data)
% At each non=IMU measurement we initialize a new node in the graph
currentPoseKey = symbol('x',measurementIndex);
currentVelKey = symbol('v',measurementIndex);
currentBiasKey = symbol('b',measurementIndex);
t = GPS_data(measurementIndex, 1).Time;
if measurementIndex == firstGPSPose
%% Create initial estimate and prior on initial pose, velocity, and biases
newValues.insert(currentPoseKey, currentPoseGlobal);
newValues.insert(currentVelKey, currentVelocityGlobal);
newValues.insert(currentBiasKey, currentBias);
newFactors.add(PriorFactorPose3(currentPoseKey, currentPoseGlobal, sigma_init_x));
newFactors.add(PriorFactorLieVector(currentVelKey, currentVelocityGlobal, sigma_init_v));
newFactors.add(PriorFactorConstantBias(currentBiasKey, currentBias, sigma_init_b));
else
t_previous = GPS_data(measurementIndex-1, 1).Time;
%% Summarize IMU data between the previous GPS measurement and now
IMUindices = find(IMUtimes >= t_previous & IMUtimes <= t);
currentSummarizedMeasurement = gtsam.ImuFactorPreintegratedMeasurements( ...
currentBias, IMU_metadata.AccelerometerSigma.^2 * eye(3), ...
IMU_metadata.GyroscopeSigma.^2 * eye(3), IMU_metadata.IntegrationSigma.^2 * eye(3));
for imuIndex = IMUindices
accMeas = [ IMU_data(imuIndex).accelX; IMU_data(imuIndex).accelY; IMU_data(imuIndex).accelZ ];
omegaMeas = [ IMU_data(imuIndex).omegaX; IMU_data(imuIndex).omegaY; IMU_data(imuIndex).omegaZ ];
deltaT = IMU_data(imuIndex).dt;
currentSummarizedMeasurement.integrateMeasurement(accMeas, omegaMeas, deltaT);
end
% Create IMU factor
newFactors.add(ImuFactor( ...
currentPoseKey-1, currentVelKey-1, ...
currentPoseKey, currentVelKey, ...
currentBiasKey, currentSummarizedMeasurement, g, w_coriolis));
% Bias evolution as given in the IMU metadata
newFactors.add(BetweenFactorConstantBias(currentBiasKey-1, currentBiasKey, imuBias.ConstantBias(zeros(3,1), zeros(3,1)), ...
noiseModel.Diagonal.Sigmas(sqrt(numel(IMUindices)) * sigma_between_b)));
% Create GPS factor
GPSPose = Pose3(currentPoseGlobal.rotation, GPS_data(measurementIndex).Position);
if mod(measurementIndex, GPSskip) == 0
newFactors.add(PriorFactorPose3(currentPoseKey, GPSPose, noiseModelGPS));
end
% Add initial value
newValues.insert(currentPoseKey, GPSPose);
newValues.insert(currentVelKey, currentVelocityGlobal);
newValues.insert(currentBiasKey, currentBias);
% Update solver
% =======================================================================
% We accumulate 2*GPSskip GPS measurements before updating the solver at
% first so that the heading becomes observable.
if measurementIndex > firstGPSPose + 2*GPSskip
isam.update(newFactors, newValues);
newFactors = NonlinearFactorGraph;
newValues = Values;
if rem(measurementIndex,10)==0 % plot every 10 time steps
cla;
plot3DTrajectory(isam.calculateEstimate, 'g-');
title('Estimated trajectory using ISAM2 (IMU+GPS)')
xlabel('[m]')
ylabel('[m]')
zlabel('[m]')
axis equal
drawnow;
end
% =======================================================================
currentPoseGlobal = isam.calculateEstimate(currentPoseKey);
currentVelocityGlobal = isam.calculateEstimate(currentVelKey);
currentBias = isam.calculateEstimate(currentBiasKey);
end
end
end % end main loop
disp('-- Reached end of sensor data')

126
matlab/gtsam_examples/IMUKittiExampleSimple.m
View File

@ -1,126 +0,0 @@
%close all
%clc
import gtsam.*;
%% Read data
IMU_metadata = importdata(gtsam.findExampleDataFile('KittiEquivBiasedImu_metadata.txt'));
IMU_data = importdata(gtsam.findExampleDataFile('KittiEquivBiasedImu.txt'));
% Make text file column headers into struct fields
IMU_metadata = cell2struct(num2cell(IMU_metadata.data), IMU_metadata.colheaders, 2);
IMU_data = cell2struct(num2cell(IMU_data.data), IMU_data.colheaders, 2);
GPS_metadata = importdata(gtsam.findExampleDataFile('KittiGps_metadata.txt'));
GPS_data = importdata(gtsam.findExampleDataFile('KittiGps.txt'));
% Make text file column headers into struct fields
GPS_metadata = cell2struct(num2cell(GPS_metadata.data), GPS_metadata.colheaders, 2);
GPS_data = cell2struct(num2cell(GPS_data.data), GPS_data.colheaders, 2);
%% Convert GPS from lat/long to meters
[ x, y, ~ ] = deg2utm( [GPS_data.Latitude], [GPS_data.Longitude] );
for i = 1:numel(x)
GPS_data(i).Position = gtsam.Point3(x(i), y(i), GPS_data(i).Altitude);
end
% % Calculate GPS sigma in meters
% [ xSig, ySig, ~ ] = deg2utm( [GPS_data.Latitude] + [GPS_data.PositionSigma], ...
% [GPS_data.Longitude] + [GPS_data.PositionSigma]);
% xSig = xSig - x;
% ySig = ySig - y;
%% Start at time of first GPS measurement
firstGPSPose = 2;
%% Get initial conditions for the estimated trajectory
currentPoseGlobal = Pose3(Rot3, GPS_data(firstGPSPose).Position); % initial pose is the reference frame (navigation frame)
currentVelocityGlobal = LieVector([0;0;0]); % the vehicle is stationary at the beginning
currentBias = imuBias.ConstantBias(zeros(3,1), zeros(3,1));
%% Solver object
isamParams = ISAM2Params;
isamParams.setFactorization('QR');
isamParams.setRelinearizeSkip(1);
isam = gtsam.ISAM2(isamParams);
newFactors = NonlinearFactorGraph;
newValues = Values;
%% Create initial estimate and prior on initial pose, velocity, and biases
newValues.insert(symbol('x',firstGPSPose), currentPoseGlobal);
newValues.insert(symbol('v',firstGPSPose), currentVelocityGlobal);
newValues.insert(symbol('b',1), currentBias);
sigma_init_x = noiseModel.Diagonal.Precisions([0;0;0; 1;1;1]);
sigma_init_v = noiseModel.Isotropic.Sigma(3, 1000.0);
sigma_init_b = noiseModel.Isotropic.Sigma(6, 0.01);
newFactors.add(PriorFactorPose3(symbol('x',firstGPSPose), currentPoseGlobal, sigma_init_x));
newFactors.add(PriorFactorLieVector(symbol('v',firstGPSPose), currentVelocityGlobal, sigma_init_v));
newFactors.add(PriorFactorConstantBias(symbol('b',1), currentBias, sigma_init_b));
%% Main loop:
% (1) we read the measurements
% (2) we create the corresponding factors in the graph
% (3) we solve the graph to obtain and optimal estimate of robot trajectory
for poseIndex = firstGPSPose:length(GPS_data)
% At each non=IMU measurement we initialize a new node in the graph
currentPoseKey = symbol('x',poseIndex);
currentVelKey = symbol('v',poseIndex);
currentBiasKey = symbol('b',1);
if poseIndex > firstGPSPose
% Summarize IMU data between the previous GPS measurement and now
IMUindices = find([IMU_data.Time] > GPS_data(poseIndex-1).Time ...
& [IMU_data.Time] <= GPS_data(poseIndex).Time);
currentSummarizedMeasurement = gtsam.ImuFactorPreintegratedMeasurements( ...
currentBias, IMU_metadata.AccelerometerSigma.^2 * eye(3), ...
IMU_metadata.GyroscopeSigma.^2 * eye(3), IMU_metadata.IntegrationSigma.^2 * eye(3));
for imuIndex = IMUindices
accMeas = [ IMU_data(imuIndex).accelX; IMU_data(imuIndex).accelY; IMU_data(imuIndex).accelZ ];
omegaMeas = [ IMU_data(imuIndex).omegaX; IMU_data(imuIndex).omegaY; IMU_data(imuIndex).omegaZ ];
deltaT = IMU_data(imuIndex).dt;
currentSummarizedMeasurement.integrateMeasurement(accMeas, omegaMeas, deltaT);
end
% Create IMU factor
newFactors.add(ImuFactor( ...
currentPoseKey-1, currentVelKey-1, ...
currentPoseKey, currentVelKey, ...
currentBiasKey, currentSummarizedMeasurement, [0;0;-9.8], [0;0;0]));
% Create GPS factor
newFactors.add(PriorFactorPose3(currentPoseKey, Pose3(currentPoseGlobal.rotation, GPS_data(poseIndex).Position), ...
noiseModel.Diagonal.Precisions([ zeros(3,1); 1./(GPS_data(poseIndex).PositionSigma).^2*ones(3,1) ])));
% Add initial value
newValues.insert(currentPoseKey, Pose3(currentPoseGlobal.rotation, GPS_data(poseIndex).Position));
newValues.insert(currentVelKey, currentVelocityGlobal);
%newValues.insert(currentBiasKey, currentBias);
% Update solver
% =======================================================================
isam.update(newFactors, newValues);
newFactors = NonlinearFactorGraph;
newValues = Values;
cla;
plot3DTrajectory(isam.calculateEstimate, 'g-');
drawnow;
% =======================================================================
currentPoseGlobal = isam.calculateEstimate(currentPoseKey);
currentVelocityGlobal = isam.calculateEstimate(currentVelKey);
currentBias = isam.calculateEstimate(currentBiasKey);
end
end
disp('TODO: display results')
% figure(1)
% hold on;
% plot(positions(1,:), positions(2,:), '-b');
% plot3DTrajectory(isam.calculateEstimate, 'g-');
% axis equal;
% legend('true trajectory', 'traj integrated in body', 'traj integrated in nav')

89
matlab/gtsam_examples/IMUKittiExampleVO.m
View File

@ -2,9 +2,11 @@ close all
clc clc
import gtsam.*; import gtsam.*;
disp('Example of application of ISAM2 for visual-inertial navigation on the KITTI VISION BENCHMARK SUITE (http://www.computervisiononline.com/dataset/kitti-vision-benchmark-suite)')
%% Read metadata and compute relative sensor pose transforms %% Read metadata and compute relative sensor pose transforms
% IMU metadata % IMU metadata
disp('-- Reading sensor metadata')
IMU_metadata = importdata('KittiEquivBiasedImu_metadata.txt'); IMU_metadata = importdata('KittiEquivBiasedImu_metadata.txt');
IMU_metadata = cell2struct(num2cell(IMU_metadata.data), IMU_metadata.colheaders, 2); IMU_metadata = cell2struct(num2cell(IMU_metadata.data), IMU_metadata.colheaders, 2);
IMUinBody = Pose3.Expmap([IMU_metadata.BodyPtx; IMU_metadata.BodyPty; IMU_metadata.BodyPtz; IMUinBody = Pose3.Expmap([IMU_metadata.BodyPtx; IMU_metadata.BodyPty; IMU_metadata.BodyPtz;
@ -20,25 +22,13 @@ VOinBody = Pose3.Expmap([VO_metadata.BodyPtx; VO_metadata.BodyPty; VO_metadata.B
VO_metadata.BodyPrx; VO_metadata.BodyPry; VO_metadata.BodyPrz; ]); VO_metadata.BodyPrx; VO_metadata.BodyPry; VO_metadata.BodyPrz; ]);
VOinIMU = IMUinBody.inverse().compose(VOinBody); VOinIMU = IMUinBody.inverse().compose(VOinBody);
% GPS metadata
GPS_metadata = importdata('KittiGps_metadata.txt');
GPS_metadata = cell2struct(num2cell(GPS_metadata.data), GPS_metadata.colheaders, 2);
GPSinBody = Pose3.Expmap([GPS_metadata.BodyPtx; GPS_metadata.BodyPty; GPS_metadata.BodyPtz;
GPS_metadata.BodyPrx; GPS_metadata.BodyPry; GPS_metadata.BodyPrz; ]);
GPSinIMU = IMUinBody.inverse().compose(GPSinBody);
%% Read data and change coordinate frame of GPS and VO measurements to IMU frame %% Read data and change coordinate frame of GPS and VO measurements to IMU frame
disp('-- Reading sensor data from file')
% IMU data % IMU data
IMU_data = importdata('KittiEquivBiasedImu.txt'); IMU_data = importdata('KittiEquivBiasedImu.txt');
IMU_data = cell2struct(num2cell(IMU_data.data), IMU_data.colheaders, 2); IMU_data = cell2struct(num2cell(IMU_data.data), IMU_data.colheaders, 2);
imum = cellfun(@(x) x', num2cell([ [IMU_data.accelX]' [IMU_data.accelY]' [IMU_data.accelZ]' [IMU_data.omegaX]' [IMU_data.omegaY]' [IMU_data.omegaZ]' ], 2), 'UniformOutput', false); imum = cellfun(@(x) x', num2cell([ [IMU_data.accelX]' [IMU_data.accelY]' [IMU_data.accelZ]' [IMU_data.omegaX]' [IMU_data.omegaY]' [IMU_data.omegaZ]' ], 2), 'UniformOutput', false);
[IMU_data.acc_omega] = deal(imum{:}); [IMU_data.acc_omega] = deal(imum{:});
%IMU_data = rmfield(IMU_data, { 'accelX' 'accelY' 'accelZ' 'omegaX' 'omegaY' 'omegaZ' });
sigma_init_x = noiseModel.Diagonal.Precisions([1;1;1; 1;1;1]);
sigma_init_v = noiseModel.Isotropic.Sigma(3, 1000.0);
sigma_init_b = noiseModel.Isotropic.Sigma(6, 0.01);
g = [0;0;-9.8];
w_coriolis = [0;0;0];
clear imum clear imum
% VO data % VO data
@ -54,32 +44,21 @@ VO_data = rmfield(VO_data, { 'dtx' 'dty' 'dtz' 'drx' 'dry' 'drz' });
noiseModelVO = noiseModel.Diagonal.Sigmas([ VO_metadata.RotationSigma * [1;1;1]; VO_metadata.TranslationSigma * [1;1;1] ]); noiseModelVO = noiseModel.Diagonal.Sigmas([ VO_metadata.RotationSigma * [1;1;1]; VO_metadata.TranslationSigma * [1;1;1] ]);
clear logposes relposes clear logposes relposes
% GPS data
GPS_data = importdata('KittiGps.txt');
GPS_data = cell2struct(num2cell(GPS_data.data), GPS_data.colheaders, 2);
% Convert GPS from lat/long to meters
[ x, y, ~ ] = deg2utm( [GPS_data.Latitude], [GPS_data.Longitude] );
for i = 1:numel(x)
GPS_data(i).Position = gtsam.Point3(x(i), y(i), GPS_data(i).Altitude);
end
% % Calculate GPS sigma in meters
% [ xSig, ySig, ~ ] = deg2utm( [GPS_data.Latitude] + [GPS_data.PositionSigma], ...
% [GPS_data.Longitude] + [GPS_data.PositionSigma]);
% xSig = xSig - x;
% ySig = ySig - y;
%% Start at time of first GPS measurement
% firstGPSPose = 1;
%% Get initial conditions for the estimated trajectory %% Get initial conditions for the estimated trajectory
% currentPoseGlobal = Pose3(Rot3, GPS_data(firstGPSPose).Position); % initial pose is the reference frame (navigation frame)
currentPoseGlobal = Pose3; currentPoseGlobal = Pose3;
currentVelocityGlobal = LieVector([0;0;0]); % the vehicle is stationary at the beginning currentVelocityGlobal = LieVector([0;0;0]); % the vehicle is stationary at the beginning
currentBias = imuBias.ConstantBias(zeros(3,1), zeros(3,1)); currentBias = imuBias.ConstantBias(zeros(3,1), zeros(3,1));
sigma_init_x = noiseModel.Isotropic.Sigmas([ 1.0; 1.0; 0.01; 0.01; 0.01; 0.01 ]);
sigma_init_v = noiseModel.Isotropic.Sigma(3, 1000.0);
sigma_init_b = noiseModel.Isotropic.Sigmas([ 0.100; 0.100; 0.100; 5.00e-05; 5.00e-05; 5.00e-05 ]);
sigma_between_b = [ IMU_metadata.AccelerometerBiasSigma * ones(3,1); IMU_metadata.GyroscopeBiasSigma * ones(3,1) ];
g = [0;0;-9.8];
w_coriolis = [0;0;0];
%% Solver object %% Solver object
isamParams = ISAM2Params; isamParams = ISAM2Params;
isamParams.setFactorization('QR'); isamParams.setFactorization('CHOLESKY');
isamParams.setRelinearizeSkip(1); isamParams.setRelinearizeSkip(10);
isam = gtsam.ISAM2(isamParams); isam = gtsam.ISAM2(isamParams);
newFactors = NonlinearFactorGraph; newFactors = NonlinearFactorGraph;
newValues = Values; newValues = Values;
@ -88,13 +67,12 @@ newValues = Values;
% (1) we read the measurements % (1) we read the measurements
% (2) we create the corresponding factors in the graph % (2) we create the corresponding factors in the graph
% (3) we solve the graph to obtain and optimal estimate of robot trajectory % (3) we solve the graph to obtain and optimal estimate of robot trajectory
timestamps = sortrows( [ ... timestamps = [VO_data.Time]';
[VO_data.Time]' 1*ones(length([VO_data.Time]), 1); ...
%[GPS_data.Time]' 2*ones(length([GPS_data.Time]), 1); ...
], 1); % this are the time-stamps at which we want to initialize a new node in the graph
timestamps = timestamps(15:end,:); % there seem to be issues with the initial IMU measurements timestamps = timestamps(15:end,:); % there seem to be issues with the initial IMU measurements
VOPoseKeys = []; % here we store the keys of the poses involved in VO (between) factors IMUtimes = [IMU_data.Time];
disp('-- Starting main loop: inference is performed at each time step, but we plot trajectory every 100 steps')
for measurementIndex = 1:length(timestamps) for measurementIndex = 1:length(timestamps)
@ -103,12 +81,6 @@ for measurementIndex = 1:length(timestamps)
currentVelKey = symbol('v',measurementIndex); currentVelKey = symbol('v',measurementIndex);
currentBiasKey = symbol('b',measurementIndex); currentBiasKey = symbol('b',measurementIndex);
t = timestamps(measurementIndex, 1); t = timestamps(measurementIndex, 1);
type = timestamps(measurementIndex, 2);
%% bookkeeping
if type == 1 % we store the keys corresponding to VO measurements
VOPoseKeys = [VOPoseKeys; currentPoseKey];
end
if measurementIndex == 1 if measurementIndex == 1
%% Create initial estimate and prior on initial pose, velocity, and biases %% Create initial estimate and prior on initial pose, velocity, and biases
@ -121,9 +93,8 @@ for measurementIndex = 1:length(timestamps)
else else
t_previous = timestamps(measurementIndex-1, 1); t_previous = timestamps(measurementIndex-1, 1);
%% Summarize IMU data between the previous GPS measurement and now %% Summarize IMU data between the previous GPS measurement and now
IMUindices = find([IMU_data.Time] >= t_previous & [IMU_data.Time] <= t); IMUindices = find(IMUtimes >= t_previous & IMUtimes <= t);
if ~isempty(IMUindices) % if there are IMU measurements to integrate
currentSummarizedMeasurement = gtsam.ImuFactorPreintegratedMeasurements( ... currentSummarizedMeasurement = gtsam.ImuFactorPreintegratedMeasurements( ...
currentBias, IMU_metadata.AccelerometerSigma.^2 * eye(3), ... currentBias, IMU_metadata.AccelerometerSigma.^2 * eye(3), ...
IMU_metadata.GyroscopeSigma.^2 * eye(3), IMU_metadata.IntegrationSigma.^2 * eye(3)); IMU_metadata.GyroscopeSigma.^2 * eye(3), IMU_metadata.IntegrationSigma.^2 * eye(3));
@ -141,28 +112,16 @@ for measurementIndex = 1:length(timestamps)
currentPoseKey, currentVelKey, ... currentPoseKey, currentVelKey, ...
currentBiasKey, currentSummarizedMeasurement, g, w_coriolis)); currentBiasKey, currentSummarizedMeasurement, g, w_coriolis));
else % if there are no IMU measurements
error('no IMU measurements in [t_previous, t]')
end
% LC: sigma_init_b is wrong: this should be some uncertainty on bias evolution given in the IMU metadata % LC: sigma_init_b is wrong: this should be some uncertainty on bias evolution given in the IMU metadata
newFactors.add(BetweenFactorConstantBias(currentBiasKey-1, currentBiasKey, imuBias.ConstantBias(zeros(3,1), zeros(3,1)), sigma_init_b)); newFactors.add(BetweenFactorConstantBias(currentBiasKey-1, currentBiasKey, imuBias.ConstantBias(zeros(3,1), zeros(3,1)), ...
noiseModel.Diagonal.Sigmas(sqrt(numel(IMUindices)) * sigma_between_b)));
%% Create GPS factor
if type == 2
newFactors.add(PriorFactorPose3(currentPoseKey, Pose3(currentPoseGlobal.rotation, GPS_data(measurementIndex).Position), ...
noiseModel.Diagonal.Precisions([ zeros(3,1); 1./(GPS_data(measurementIndex).PositionSigma).^2*ones(3,1) ])));
end
%% Create VO factor %% Create VO factor
if type == 1
VOpose = VO_data(measurementIndex).RelativePose; VOpose = VO_data(measurementIndex).RelativePose;
newFactors.add(BetweenFactorPose3(VOPoseKeys(end-1), VOPoseKeys(end), VOpose, noiseModelVO)); newFactors.add(BetweenFactorPose3(currentPoseKey - 1, currentPoseKey, VOpose, noiseModelVO));
end
% Add initial value % Add initial value
% newValues.insert(currentPoseKey, Pose3(currentPoseGlobal.rotation, GPS_data(measurementIndex).Position)); newValues.insert(currentPoseKey, currentPoseGlobal.compose(VOpose));
newValues.insert(currentPoseKey,currentPoseGlobal);
newValues.insert(currentVelKey, currentVelocityGlobal); newValues.insert(currentVelKey, currentVelocityGlobal);
newValues.insert(currentBiasKey, currentBias); newValues.insert(currentBiasKey, currentBias);
@ -172,9 +131,13 @@ for measurementIndex = 1:length(timestamps)
newFactors = NonlinearFactorGraph; newFactors = NonlinearFactorGraph;
newValues = Values; newValues = Values;
if rem(measurementIndex,20)==0 % plot every 20 time steps if rem(measurementIndex,100)==0 % plot every 100 time steps
cla; cla;
plot3DTrajectory(isam.calculateEstimate, 'g-'); plot3DTrajectory(isam.calculateEstimate, 'g-');
title('Estimated trajectory using ISAM2 (IMU+VO)')
xlabel('[m]')
ylabel('[m]')
zlabel('[m]')
axis equal axis equal
drawnow; drawnow;
end end
@ -185,3 +148,5 @@ for measurementIndex = 1:length(timestamps)
end end
end % end main loop end % end main loop
disp('-- Reached end of sensor data')