Risk Assessment - Henrrrrrry/ART-x3-25s1 GitHub Wiki

🔍 Risk Assessment for Human Tracking Camera Project

1. Motor Doesn’t Rotate This is one of the most critical risks in the project. If the motor fails to rotate, the camera cannot pan or tilt, directly hindering the system’s ability to track humans. The likelihood of occurrence is moderate, but its impact is severe, potentially halting progress entirely.
Probability: 3 / Impact: 5

2. Hardware Performance Limitation This risk concerns whether the chosen hardware (OpenMV H7) can sustain the performance required for human tracking. Due to its limited processing power compared to alternatives like Raspberry Pi, there is a considerable chance it may not maintain the necessary frame rate or model accuracy.
Probability: 4 / Impact: 3

3. Hardware Failure This includes damage to components such as the camera, motor, or circuit elements. While critical components are generally reliable, wiring and connectors may fail due to electrical issues or physical stress. The chance is relatively low, but consequences are significant.
Probability: 2 / Impact: 5

4. Data or Code Loss This refers to accidental deletion or corruption of code or data, potentially caused by the IDE or human error. Although possible, the team follows regular manual backup practices, reducing this risk considerably.
Probability: 1 / Impact: 5

5. Team Member Conflict Conflicts within the team can affect productivity and project coordination. However, the team currently has a positive and cooperative atmosphere, making the likelihood of serious internal issues quite low.
Probability: 1 / Impact: 3

6. Conflict Between Team and Client Miscommunication or mismatched expectations with the client can lead to misalignment in deliverables. The team plans to maintain regular contact with the client to clarify requirements and expectations, mitigating this risk.
Probability: 1 / Impact: 4

7. Privacy Violation As the camera tracks human subjects, it may raise privacy concerns, especially in public environments. While privacy principles will be respected, the perception of surveillance could still affect public acceptance.
Probability: 2 / Impact: 4

8. Lighting Interference The accuracy of object detection may decrease under poor or variable lighting conditions, such as low light or backlighting. Since environmental light cannot always be controlled, this poses a moderate risk to tracking performance.
Probability: 3 / Impact: 3

9. Short Circuit or Wiring Errors Incorrect wiring or short circuits could damage hardware and disrupt operation, especially during the assembly stage. While preventable with care, the possibility still exists.
Probability: 2 / Impact: 4

10. Motor Overheating Continuous operation or abnormal load may cause the motor to overheat, reducing lifespan or creating safety hazards. Although mechanisms for control are planned, the risk remains under high-stress use.
Probability: 2 / Impact: 4

📉 Risk Assessment Matrix (Impact from Low to High)

Impact ↓ / Probability → 1 (Rare) 2 (Unlikely) 3 (Possible) 4 (Likely) 5 (Almost Certain)
1 (Negligible)
2 (Minor)
3 (Moderate) 5 8 2
4 (Major) 6 9, 10
5 (Severe) 4 3 1