get
PURPOSE 
Get function for trajectories.
SYNOPSIS
function out=get(T,field)
DESCRIPTION
Get function for trajectories.
Retrives data from a trajectory. The possible fields to retrive are
- type: Type of trajectory (line, ellipse, mixed).
- data: The cell array of poses defining the trajectory.
- averageDisplacement: The average displacement between poses. This
is used to compute the amount of error to use when corrupting the
ground truth trajectories. For example, odometry sensors usually
produce an error that is aboud 5% of the displacement. The average
displacement is used to define a constant error for the odometry.
- boundingBox: The interval vector bounding the area where the
trajectory is defined. This is used to define the area where to
robot is supposed to move. This is sometimes used to plot.CROSS-REFERENCE INFORMATION
This function calls:
- size Number of poses in a trajectory.
- size Size of the state estimated in the filter
- Interval Interval constructor
- double Converts an interval to a double.
- size Size (rows/columns) of an interval matrix.
- Absolute2Relative Relative displacement between two poses.
- double Converts a double
- size Number of parameters of the pose.
- size Number of readings stored in the Sensor.
- Trajectory Trajectory constructor.
- Experiment1a Simulated experiment.
- Experiment1b Simulated experiment.
- Experiment1c Simulated experiment.
- GetCrossCovarianceWithPose Cross covariance of a given pose with the current one.
- GetPoseEstimation Estimation of a given pose.
- P_EKF Constructor for Extended Kalman Filter for PoseSLAM.
- Update Adds a new pose to an EKF filter.
- get Get function for EKF filters.
- EKF_Prediction EKF Prediction step.
- EKF_Update EKF Update step.
- ComputeNEES Quantifies the consitency of a filter.
- GetMarginalCovariance Marginal covariance between the current pose and a previous one.
- GetPoseCovariance Marginal covariance of a given pose.
- GetPoseData Defines a PoseData objects from the information stored in the filter.
- GetPoseMean Mean estimation of a given pose.
- InitFilterPlot Initializes the plot of a filter.
- P_Filter Construtor for the base Pose SLAM filter.
- ProbDistance2PoseBelowThreshold Checks if pose 'step' is close enought to the current pose.
- Simulation Simulates a robot performing Pose SLAM.
- TreeDetectLoops Propose loops based on the distance between poses using a tree for the search.
- UpdateFilterPlot Updates the plot of a filter with a new pose.
- UpdateLoop Adds a loop to the filter.
- get Generic get function for filters.
- GetCrossCovarianceWithPose Cross covariance of a given pose with the current one.
- GetPoseCovariance Covariance estimation of a given pose.
- GetPoseEstimation Estimation of a given pose.
- GetPoseMean Mean estimation of a given pose.
- P_TRO Constructor of TRO Filters for Pose SLAM.
- Update Adds a new pose to a TRO filter.
- UpdateLoop Adds a new loop to a TRO filter.
- get Get function for TRO filters.
- TRO_Prediction Prediction Step.
- TRO_Update Filter update step.
- InsertInTree Inserts an element in a BTree.
- SearchInTree Search nearest-neighbours using a Btree
- UpdateTree Updates the information in a BTree.
- ProbDistance2PoseDataAboveThreshold p=ProbDistance2PoseDataAboveThreshold(PD,F,t,pt)
- PlotGaussian Plots a Gaussian.
- PlotPoseSequence Plots a sequence of poses.
- ContinueSimulation Continues a previous partial simulation.
- char Converts a pose to a string.
- PlotSensor Plots a sensors.
- get Generic get for relative positioning sensors
- SensorLoopCloses Checks if two sensor readigs can be properly registered.
SOURCE CODE
0001 function out=get(T,field) 0002 % Get function for trajectories. 0003 % 0004 % Retrives data from a trajectory. The possible fields to retrive are 0005 % - type: Type of trajectory (line, ellipse, mixed). 0006 % - data: The cell array of poses defining the trajectory. 0007 % - averageDisplacement: The average displacement between poses. This 0008 % is used to compute the amount of error to use when corrupting the 0009 % ground truth trajectories. For example, odometry sensors usually 0010 % produce an error that is aboud 5% of the displacement. The average 0011 % displacement is used to define a constant error for the odometry. 0012 % - boundingBox: The interval vector bounding the area where the 0013 % trajectory is defined. This is used to define the area where to 0014 % robot is supposed to move. This is sometimes used to plot. 0015 0016 switch field 0017 case 'type' 0018 out=T.type; 0019 case 'data' 0020 out=T.data; 0021 case 'averageDisplacement' 0022 n=size(T.data,2); 0023 d=zeros(size(T.data{1}),1); 0024 for i=2:n 0025 d=d+abs(double(Absolute2Relative(T.data{i-1},T.data{i}))); 0026 end 0027 out=d/(n-1); 0028 case 'boundingBox' 0029 n=size(T.data,2); 0030 ma=double(T.data{1}); 0031 mi=ma; 0032 for i=2:n 0033 p=double(T.data{i}); 0034 ma=max(ma,p); 0035 mi=min(mi,p); 0036 end 0037 out=1.5*Interval(mi,ma); 0038 out(3)=Interval(-pi,pi); 0039 otherwise 0040 error('Unknow field in Trajectory get'); 0041 end