Skip to content

matlab/matlab-agentic-toolkit / skills-catalog/radar/matlab-import-tracking-data

matlab-import-tracking-data

Verified

Import raw data (CSV, XLSX, TXT, or MATLAB tables) into formats used by Sensor Fusion and Tracking Toolbox. Handles both ground truth trajectories and sensor detection data. For truth: builds trackingScenarioRecording, tuning timetable, truthlog, or converted table. For sensor data: builds task-oriented dataFormat structs (preferred) or objectDetection arrays (legacy). Use when importing flight logs, GPS logs, radar detections, IR measurements, lidar/camera bounding boxes, ADS-B data, AIS ship tracks, or any recorded data for use with trackers, filter tuning, or tracker evaluation.

references/code-patterns.md

View on GitHub
# Complete Code Patterns

## Flight Log → Scenario Recording (Geodetic, ECEF output)

```matlab
%% 1. Read data
rawTable = readtable("flight_data.csv", TextType="string", VariableNamingRule="preserve");

%% 2. Extract and convert
% Time
epochSec = rawTable.("timestamp");
dt = datetime(epochSec, 'ConvertFrom', 'posixtime');
simTime = seconds(dt - dt(1));

% Platform IDs — remap to sequential integers
rawPID = rawTable.("aircraft_id");
[~, ~, platformIDs] = unique(rawPID, 'stable');

% Position: geodetic → ECEF
lat = rawTable.("lat");  lon = rawTable.("lon");
alt = rawTable.("alt") * 0.3048;  % ft → m (adjust as needed)
[xE, yE, zE] = geodetic2ecef(wgs84Ellipsoid, lat, lon, alt);
posECEF = [xE yE zE];

% Velocity: NED → ECEF (vectorized, no Aerospace Toolbox needed)
velNED = [rawTable.("vn"), rawTable.("ve"), rawTable.("vd")];
[vxE, vyE, vzE] = ned2ecefv(velNED(:,1), velNED(:,2), velNED(:,3), lat, lon);
velECEF = [vxE vyE vzE];

% Orientation: Euler (deg) → quaternion, NED → ECEF
yaw_r = deg2rad(rawTable.("heading"));
pitch_r = deg2rad(rawTable.("pitch"));
roll_r = deg2rad(rawTable.("roll"));
qNED = quaternion([yaw_r pitch_r roll_r], 'euler', 'ZYX', 'frame');
qECEF = quaternion.zeros(height(rawTable), 1);
for ii = 1:height(rawTable)
    latR = deg2rad(lat(ii)); lonR = deg2rad(lon(ii));
    R = [-sin(latR)*cos(lonR), -sin(latR)*sin(lonR),  cos(latR);
         -sin(lonR),            cos(lonR),             0;
         -cos(latR)*cos(lonR), -cos(latR)*sin(lonR), -sin(latR)];
    qECEF(ii) = quaternion(R, 'rotmat', 'frame') * qNED(ii);
end

% Defaults for missing states
acc = zeros(height(rawTable), 3);
angVel = zeros(height(rawTable), 3);

%% 3. Sort by time
[simTime, si] = sort(simTime);
platformIDs = platformIDs(si); posECEF = posECEF(si,:);
velECEF = velECEF(si,:); acc = acc(si,:);
qECEF = qECEF(si); angVel = angVel(si,:);

%% 4. Build trackingScenarioRecording
uniqueTimes = unique(simTime, 'stable');
recordedData = struct('SimulationTime', cell(numel(uniqueTimes),1), 'Poses', []);
for ii = 1:numel(uniqueTimes)
    idx = find(simTime == uniqueTimes(ii));
    poses = repmat(struct('PlatformID',[],'ClassID',[],'Position',[], ...
        'Velocity',[],'Acceleration',[],'Orientation',[],'AngularVelocity',[]), numel(idx), 1);
    for jj = 1:numel(idx)
        k = idx(jj);
        poses(jj) = struct('PlatformID',platformIDs(k), 'ClassID',0, ...
            'Position',posECEF(k,:), 'Velocity',velECEF(k,:), ...
            'Acceleration',acc(k,:), 'Orientation',qECEF(k), ...
            'AngularVelocity',angVel(k,:));
    end
    recordedData(ii).SimulationTime = uniqueTimes(ii);
    recordedData(ii).Poses = poses;
end
tsr = trackingScenarioRecording(recordedData, ...
    CoordinateSystem="Geodetic", IsEarthCentered=true);
```

## Driving Log → Truth Log (Cartesian Scenario frame)

```matlab
%% 1. Read data
rawTable = readtable("driving_data.csv", TextType="string", VariableNamingRule="preserve");

%% 2. Extract
timeSec = rawTable.("time");
simTime = timeSec - timeSec(1);
[~, ~, platformIDs] = unique(rawTable.("object_id"), 'stable');
pos = [rawTable.("x") rawTable.("y") rawTable.("z")];
vel = [rawTable.("vx") rawTable.("vy") rawTable.("vz")];
acc = [rawTable.("ax") rawTable.("ay") rawTable.("az")];
ori = quaternion(deg2rad([rawTable.("yaw") rawTable.("pitch") rawTable.("roll")]), ...
    'euler', 'ZYX', 'frame');
angVel = zeros(height(rawTable), 3);
angAcc = zeros(height(rawTable), 3);

%% 3. Sort
[simTime, si] = sort(simTime);
platformIDs=platformIDs(si); pos=pos(si,:); vel=vel(si,:); acc=acc(si,:);
ori=ori(si); angVel=angVel(si,:); angAcc=angAcc(si,:);

%% 4. Build truth log
uniqueTimes = unique(simTime, 'stable');
truthLog = cell(numel(uniqueTimes), 1);
for ii = 1:numel(uniqueTimes)
    idx = find(simTime == uniqueTimes(ii));
    poses = repmat(struct('Time',NaN,'PlatformID',NaN,'ClassID',NaN, ...
        'Position',NaN(1,3),'Velocity',NaN(1,3),'Acceleration',NaN(1,3), ...
        'Orientation',NaN,'AngularVelocity',NaN(1,3),'AngularAcceleration',NaN(1,3)), ...
        numel(idx), 1);
    for jj = 1:numel(idx)
        k = idx(jj);
        poses(jj) = struct('Time',uniqueTimes(ii), 'PlatformID',platformIDs(k), ...
            'ClassID',0, 'Position',pos(k,:), 'Velocity',vel(k,:), ...
            'Acceleration',acc(k,:), 'Orientation',ori(k), ...
            'AngularVelocity',angVel(k,:), 'AngularAcceleration',angAcc(k,:));
    end
    truthLog{ii} = poses;
end
```

## GPS Log → Tuning Data (ECEF output)

```matlab
%% 1. Read data
rawTable = readtable("gps_log.csv", TextType="string", VariableNamingRule="preserve");

%% 2. Extract
epochSec = rawTable.("timestamp");
dt = datetime(epochSec, 'ConvertFrom', 'posixtime');

lat = rawTable.("latitude"); lon = rawTable.("longitude"); alt = rawTable.("altitude");
[xE, yE, zE] = geodetic2ecef(wgs84Ellipsoid, lat, lon, alt);
pos = [xE yE zE];

% Velocity if available, else zeros
if all(ismember(["vn","ve","vd"], rawTable.Properties.VariableNames))
    velNED = [rawTable.("vn") rawTable.("ve") rawTable.("vd")];
    [vxE, vyE, vzE] = ned2ecefv(velNED(:,1), velNED(:,2), velNED(:,3), lat, lon);
    vel = [vxE vyE vzE];
else
    vel = zeros(height(rawTable), 3);
end

if ismember("id", rawTable.Properties.VariableNames)
    [~,~,platformIDs] = unique(rawTable.("id"), 'stable');
else
    platformIDs = ones(height(rawTable), 1);
end

acc = zeros(height(rawTable), 3);

%% 3. Sort
[dt, si] = sort(dt);
platformIDs=platformIDs(si); pos=pos(si,:); vel=vel(si,:); acc=acc(si,:);

%% 4. Build tuning data (position, velocity, acceleration only)
uniquePIDs = unique(platformIDs, 'stable');
startTime = min(dt);
tuningData = cell(numel(uniquePIDs), 1);
for ii = 1:numel(uniquePIDs)
    idx = platformIDs == uniquePIDs(ii);
    Time = dt(idx) - startTime;
    tuningData{ii} = timetable(Time, ...
        repmat(uniquePIDs(ii),sum(idx),1), pos(idx,:), vel(idx,:), acc(idx,:), ...
        VariableNames=["PlatformID","Position","Velocity","Acceleration"]);
end
if isscalar(tuningData), tuningData = tuningData{1}; end
```


----

Copyright 2026 The MathWorks, Inc.
deprecation-and-migrationaddyosmani/agent-skillsManages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
performance-optimizationaddyosmani/agent-skillsOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
Agent Cost Reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
Agent Cost Reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall