Before you begin
Starting knowledge
- Basic geometry, units and programming.
- Know the intended measurement and operating environment.
Equipment & environment
Start with recorded data. Use documented mounts and supervised equipment setup for physical captures.
No fixed completion time: use the evidence gates below to decide when to advance. Resource links open the botstest directory entry; check its publisher documentation for the exact version and configuration.
Beginner
Specify what must be observed
Choose sensing by measurement needs instead of a headline specification.
Learn in this order
- Range, field of view and scene limitations
- Camera intrinsics and sensor frames
- Bandwidth, interfaces and timestamp sources
Put it into practice
- Write an observation requirement for one scene or route.
- Inspect example sensor output and mark invalid or missing observations.
A sensing requirements sheet and sample-data audit.
Ready for the next step when…
Resources for this stage
Compare depth sensing against range, scene, calibration and supported software requirements.
Compare depth-camera options against the actual observation and integration requirements.
Study LiDAR integration, data interfaces and sensing conditions for the route or scene.
Intermediate
Calibrate geometry and align time
Check whether multiple observations can be compared in one frame.
Learn in this order
- Intrinsic/extrinsic calibration and mounting stability
- Clock alignment and transport delay
- Reference targets and residual inspection
Put it into practice
- Capture a repeatable target or reference scene.
- Compare aligned camera, depth or inertial observations and inspect residual errors.
A calibration package and alignment validation report.
Ready for the next step when…
Resources for this stage
Explore fiducial detection and pose observations with known marker geometry and camera calibration.
Inspect point clouds and geometry-processing results before using them in a perception pipeline.
Explore inertial measurement and state-estimation integration with explicit calibration and timing.
Professional
Operate a monitored perception chain
Detect when calibration or sensing assumptions no longer hold.
Learn in this order
- Drift, thermal changes and scene shift
- Data-quality monitoring and release records
- Recalibration triggers and component substitution
Put it into practice
- Define checks after a mount change, sensor replacement or software update.
- Replay normal and degraded recordings through the processing pipeline.
A sensor acceptance suite and recalibration runbook.
Ready for the next step when…
Resources for this stage
Visualize spatial observations and compare them with model and trajectory outputs.
Inspect recorded robot signals and build views that expose timing and system failures.
Investigate depth and edge-perception hardware in a recorded sensing experiment.
Common pitfalls
- Confusing a product range specification with usable measurements in your scene.
- Fixing a time-alignment issue by changing geometry.
Where to go next
Continue to perception research, navigation or inspection with the validated sensing chain.
Explore other build paths →