home Home / Safety LiDAR vs. 3D Camera on AMR: Protection Functions, Blind Spots, and Verification Boundaries

Safety LiDAR vs. 3D Camera on AMR: Protection Functions, Blind Spots, and Verification Boundaries

AMR sensors serve different roles but are often conflated.

Navigation/perception sensors: Used for mapping, localization, path planning, and obstacle detection. These sensors let the robot “see” the environment and make navigation decisions. Their role in the safety architecture is product-specific. Treat any LiDAR, camera, or 3D sensor as part of a safety-related protective function only when the complete device and safety architecture are safety-rated and validated for that function.

Safety protection devices: Used for functional safety — detecting people or obstacles entering danger zones and triggering deceleration or stop. These sensors must meet functional safety standards with defined response time, detection range, and diagnostic coverage.

Core misunderstanding: Buyers see a 3D camera or LiDAR on an AMR and assume “has sensors = safe.” But if those sensors are only for navigation perception without safety certification, they cannot replace safety protection devices. An AMR with a 3D camera may “see” a person — but if seeing a person does not trigger a safety stop through a safety-certified circuit, that “seeing” has no safety meaning.


What Safety LiDAR Typically Provides

Safety LiDAR (safety LiDAR scanner) is the most common safety protection device on AMRs.

Safety LiDAR functions:

  • Scans horizontally, detects people or objects entering the protective field.
  • Sets two protective fields: warning field (detect → decelerate) and protective field (detect → stop).
  • Fields are configurable — different zones can have different shapes and responses.
  • Meets functional safety levels — the specific PL or SIL rating depends on the product and configuration.

Key parameters to confirm:

ParameterWhy it mattersHow to confirm
Detection range (protective and warning)Determines max safe speedDatasheet or test report
Resolution (smallest detectable object at range)Determines what the sensor can reliably detectDatasheet
Response time (detection to motor stop)Determines stopping distanceDatasheet + on-site test
Safety level (PL or SIL)Functional safety levelSafety certificate
Diagnostic coverageFault detection capabilitySafety certificate

Confirm the safety LiDAR’s safety level certificate (not the navigation LiDAR’s parameters), protective field configuration, and response time.


What 3D Cameras Add Beyond the Scan Plane

Safety LiDAR scans horizontally — but AMR obstacles are not all at scan-plane height.

Safety LiDAR blind spots:

  • Obstacles above the scan plane — outstretched arms, shelf beams, open cabinet doors.
  • Obstacles below the scan plane — small floor objects, step edges.
  • Transparent or highly reflective objects — glass doors, mirrors (LiDAR may misdetect or miss).

3D cameras may provide supplementary coverage:

  • May detect obstacles above the scan plane, depending on mounting position and field of view.
  • May detect low obstacles, depending on mounting angle and configuration.
  • May provide semantic understanding (distinguish “person” from “shelf”), depending on software capability.

But 3D cameras are not automatically safety-rated devices. They can provide auxiliary perception — detecting an obstacle and triggering deceleration/stop through software — but this path does not go through a functional safety circuit unless the complete device/system is safety-rated and validated for that function.

Key distinction:

  • A safety-rated protective function must detect the defined hazard and bring the vehicle to the validated safe state through the certified safety architecture. The actual safety chain may involve a safety controller, safe drive functions, relays, or other architecture defined by the OEM.
  • Treat a 3D sensor as part of a safety function only when the complete device/system is safety-rated and validated for that function.

If the software path has latency, crash, or error, the 3D camera’s detection does not trigger a safety stop.


Blind Spots for Overhead and Low Obstacles

Overhead obstacles:

  • Shelf beams above the robot but not at safety LiDAR scan height.
  • Open cabinet doors, protruding pipes.
  • Robot + load top may hit these obstacles.

Low obstacles:

  • Small floor objects (tools, debris, cables).
  • Step edges and floor height differences.
  • Objects below the safety LiDAR scan plane may be missed.

Solutions:

  • Overhead: Install 3D camera or ToF sensor for upper detection — but if it is a safety need, a safety-certified 3D sensor is required.
  • Low: Adjust safety LiDAR mounting height or add low detection sensors — again, safety-certified.
  • Physical: Mark known overhead obstacles in path planning, restrict robot entry.

Confirm what zones the AMR’s safety protection covers — only horizontal? Or including above and below? If only horizontal, do you have overhead or low obstacle risks on your routes?


Safety-Certified Functions vs. Auxiliary Perception

Sensor roleSensor typeSafety levelPost-detection responseApplicable
Safety protectionSafety LiDARPL or SIL per product certificateSafety circuit direct stopPersonnel protection
Safety protectionSafety 3D camera (if certified)PL or SIL per product certificateSafety circuit direct stopOverhead/low protection
Auxiliary perceptionNavigation LiDARNoneSoftware deceleration/obstacle avoidanceNavigation and path planning
Auxiliary perception3D cameraNoneSoftware deceleration/obstacle avoidanceObstacle recognition and classification
Auxiliary perceptionUltrasonicsNoneSoftware decelerationClose-range detection

Some 3D sensing systems are available for safety-related applications, while many 3D cameras are used only for navigation or perception. Verify the safety rating, intended function, detection limits, and validated architecture of the specific system rather than assuming the sensor type determines safety status. If a project needs 3D-level safety protection (e.g., overhead obstacles), confirm whether the supplier has a safety-certified 3D sensor solution, or use physical solutions (path restriction + isolation) instead.

Auxiliary sensors can improve AMR obstacle avoidance — making the robot “smarter.” But if your safety risk assessment requires that personnel entering the danger zone triggers a functional safety response (safety circuit stop, not software deceleration), then safety-certified sensors are mandatory.


Environmental Sensitivity and Diagnostic Coverage

Safety LiDAR environmental sensitivity:

  • Dust: High concentration may cause false detection (“false obstacles”).
  • Smoke: Scatters laser, causing signal loss or misjudgment.
  • Glass/mirrors: Reflection may cause detection range anomalies or misses.
  • Direct sunlight: Strong light may interfere with the receiver.
  • Vibration: Mounting loosening causes scan plane deviation.

Diagnostic coverage:

  • Safety LiDAR has diagnostic coverage and self-test per its safety certification — enters safe state (stop) when it detects its own fault. Confirm the specific diagnostic coverage and fault response from the safety certificate.
  • Auxiliary sensors typically lack diagnostic coverage — faults may go undetected.

If your environment has dust, smoke, glass, or strong light (e.g., welding arc), confirm safety LiDAR detection reliability under these conditions. If unreliable, additional protective measures (physical isolation, path restriction) are needed.


How to Verify Protective Fields On-Site

Safety LiDAR protective field configuration is not “set and forget” — it needs on-site verification.

Verification method:

  1. Use the test object specified by the safety LiDAR manufacturer’s test procedure.
  2. Move the test object at the protective field boundary.
  3. Confirm stop triggers at the protective field boundary.
  4. Confirm deceleration triggers at the warning field boundary.
  5. Measure response time from detection to stop.
  6. Measure stopping distance — from trigger to complete stop.
  7. Repeat at different speeds and load conditions.

Verification frequency:

  • Full verification at initial deployment.
  • Re-verify after every protective field configuration change.
  • Periodic review per your safety management plan.

Confirm the supplier provides safety LiDAR verification methods and test object specifications. Confirm who maintains and updates the protective field configuration.


Sensor Role Verification Checklist

SensorRoleSafety levelDetection rangeBlind spotsResponse methodVerification method
Safety LiDAR      
Primary safety LiDAR☐ Safety ☐ AuxiliaryPL______mAbove/below/transparentSafety circuit stopTest object verification
Secondary safety LiDAR☐ Safety ☐ AuxiliaryPL______mSafety circuit stopTest object verification
Navigation/perception      
Navigation LiDAR☐ Navigation ☐ AuxiliaryNone___mSoftware decel/avoidFunctional test
3D camera☐ AuxiliaryNone___mSoftware decel/avoidFunctional test
Ultrasonics☐ AuxiliaryNone___mSoftware decelFunctional test
Blind spot assessment      
Overhead obstacle risk☐ Yes ☐ NoRoute assessment
Low obstacle risk☐ Yes ☐ NoRoute assessment
Transparent object risk☐ Yes ☐ NoRoute assessment
Environment      
Dust/smoke☐ Yes ☐ NoEnvironmental assessment
Glass/mirrors☐ Yes ☐ NoRoute assessment
Strong light/welding arc☐ Yes ☐ NoRoute assessment
Verification      
Protective field verification methodConfirm method
Verification frequencyConfirm frequency
Configuration maintenance responsibilityConfirm owner

Need to Define Safety Sensing for an AMR Project?

Safety LiDAR and 3D sensing can play different roles depending on the certified architecture. We can help map blind spots, stopping performance, and site hazards to the required protective functions.

Please share, if available: vehicle speed and load, stopping-distance data, route geometry, blind spots, pedestrian interaction, required safety functions, and the proposed sensing architecture.

Review Safety Sensing

Safety Sensing Verification Inputs

  1. Distinguish navigation perception from safety protection — having sensors does not mean safe; safety certification is what matters.
  2. Safety LiDAR is the standard horizontal-plane protection — but has blind spots above and below.
  3. 3D cameras are auxiliary perception, not safety protection — unless they are safety-certified 3D sensors.
  4. Assess overhead and low obstacle risks — if your routes have these, additional protection is needed.
  5. Environmental sensitivity affects safety LiDAR reliability — dust, smoke, glass, and strong light can cause false detection or misses.
  6. Protective field verification is a deployment must — not set-and-forget; needs periodic review per your safety management plan.
  7. Safety level depends on the specific product and configuration — do not assume a universal PL or SIL rating.

Contact Us