Capstone & Resources · Study deck
Capstone Projects: Selection and Fleet Tracking
A student can buy every sensor and still have no project that can be reviewed.
Sammy is your guide for this deck.
After studying this chapter
Learning objectives
You will be able to:
- Select a capstone scope from skill, evidence, and integration risk.
- Build a claim-to-evidence table and a fleet-tracking test plan.
- Explain: A student can buy every sensor and still have no project that can be reviewed.
- Explain: Each brief below is a complete, independent project -- pick the one that matches your experience level and interests, then open its page for the full requirements, hardware list, architecture, milestones, and rubric.
Major section
Start With the Decision
A student can buy every sensor and still have no project that can be reviewed.
- The smallest useful claim, evidence table, and integration test must come before assembly.
Major section
Choose Your Project
Each brief below is a complete, independent project -- pick the one that matches your experience level and interests, then open its page for the full requirements, hardware list, architecture, milestones, and rubric.
- All three share the reviewable-claim approach above.
Major section
Project Submission Guidelines
Step 1 — Technical Report (10-15 pages) Problem statement and requirements Architecture design with diagrams Implementation details Testing results and analysis Lessons learned.
- Step 4 — Hardware Documentation: Bill of materials with costs Wiring diagrams Assembly photos.
Major section
Try It: Build a Claim-to-Evidence Table
If a claim has no evidence, either gather the evidence before the demo or remove the claim from the presentation.
- Cover: problem, solution, demo, results, lessons.
Calibration note, datasheet tolerance, or test log.
The audience cannot tell whether readings are accurate or merely displayed.
Timestamped sensor-to-dashboard trace.
A dashboard screenshot could hide gateway, broker, or storage failures.
Wiring diagram, setup steps, and known-limitations note.
The next maintainer may need to reverse-engineer the prototype before fixing it.
Major section
Decision Framework: Capstone Project Selection by Experience Level
Project requires three or more technologies the team has never used.
- Budget is more than $200 per team member.
- Timeline is compressed below four weeks for a moderate-complexity system.
- Core hardware has no reliable examples, documentation, or backup option.
- No fallback plan exists if the primary approach fails.
Major section
Common Mistake: Scope Creep Kills Capstone Projects
The Mistake: Team starts with "Environment Monitor" (4-week project).
- Week 2: "Let's add actuators!" Week 4: "What about solar power?" Week 6: "Should we use LoRa instead of Wi-Fi?" Week 8: Project incomplete, demo is half-working prototype.
Major section
Concept Relationships
Budget constraints force architecture decisions: A $120 budget might eliminate LoRa and solar in the first version, pushing the team toward Wi-Fi, USB charging, and a stricter must-have feature list.
- Protocol selection from networking modules helps compare MQTT, LoRa, and cellular trade-offs.
Major section
3. Not planning for hardware failure and component replacement
Design your system to continue operating (degraded mode) when individual sensors fail.
- Capstone projects that depend on every component working perfectly fail during live demonstrations.
Major section
Capstone Project 3: Fleet Tracking System
The shared capstone approach is introduced on the Capstone Projects hub, which also has the submission guidelines and the project-selection decision framework; this part works through the Fleet Tracking System brief.
- A moving dot alone cannot support an investigation.
- Telemetry means measurements and status sent from a device for review elsewhere.
Major section
Capstone Project 3: Fleet Tracking System (continued)
This exceeds the 30-day target by about 5.8x.
- The deeper project adds architecture, power, storage, alerts, privacy, and release evidence.
- The system should provide real-time location tracking, geofence alerts, movement history, and battery-efficient operation for long deployment without charging.
- Each stage must preserve the same vehicle identity and timestamp contract.
Major section
Capstone Project 3: Fleet Tracking System (continued)
Step 3 — Backend stores the point, checks geofence transitions, and updates the dashboard.
- Your system should focus on reliable location updates, safe geofence logic, and power-aware reporting rather than constant high-rate tracking.
- With an 80% real-world efficiency allowance, the numerical estimate still exceeds 30 days, but measure cellular reconnection and GPS acquisition energy before claiming that margin in the field.
- A measured trace must validate that estimate.
Major section
Capstone Project 3: Fleet Tracking System (continued)
A strong score therefore depends on the connections between criteria, not isolated demonstrations.
- Step 2 — Firmware obtains a GPS fix, packages location and battery status, and transmits over LTE-M or NB-IoT.
- Step 4 — Operators review alerts, routes, and device health from the web interface.
- The must-have path combines GPS location, LTE-M or NB-IoT connectivity, live map state, route history, geofence alerts, battery level, and documentation.
Deck summary
Key takeaways
A student can buy every sensor and still have no project that can be reviewed.
- Each brief below is a complete, independent project -- pick the one that matches your experience level and interests, then open its page for the full requirements, hardware list, architecture, milestones, and rubric.
- Step 1 — Technical Report (10-15 pages) Problem statement and requirements Architecture design with diagrams Implementation details Testing results and analysis Lessons learned.
- If a claim has no evidence, either gather the evidence before the demo or remove the claim from the presentation.
- Project requires three or more technologies the team has never used.
Retrieval practice
Recall check 1 of 6
Test Tessa says: answer from memory, then check your reasoning.
Q1You're starting an IoT capstone project with a 6-week timeline. During week 1, you discover that the specialized sensor you planned to use has a 4-week lead time. What is the BEST approach?
Show answer
Answer: A Flexibility in component selection is crucial.
Retrieval practice
Recall check 2 of 6
Test Tessa says: answer from memory, then check your reasoning.
Q2For a battery-powered MQTT sensor node that sends sparse readings and has no Linux workload, why might a team choose ESP32 over Raspberry Pi?
Show answer
Answer: A This answer demonstrates understanding of: (1) power consumption trade-offs, (2) matching platform capabilities to requirements, (3) not over-engineering the solution.
Retrieval practice
Recall check 3 of 6
Test Tessa says: answer from memory, then check your reasoning.
Q3What proves that sensing, transport, processing and presentation interfaces work together?
Show answer
Answer: C System integration proves the interfaces between sensing, transport, processing, and presentation instead of treating isolated component demos as a working project.
Retrieval practice
Recall check 4 of 6
Test Tessa says: answer from memory, then check your reasoning.
Q4The Fleet Tracking System uses LTE-M or NB-IoT rather than standard 4G LTE. For a tracker sending 100-byte location updates once per minute, why are these IoT-specific cellular technologies preferred?
Show answer
Answer: B LTE-M and NB-IoT suit small telemetry reports and can use PSM/eDRX to sleep between reports, saving energy.
Retrieval practice
Recall check 5 of 6
Test Tessa says: answer from memory, then check your reasoning.
Q5The power optimization code uses an accelerometer to detect movement and adjust the GPS reporting interval. Why is movement detection essential rather than just using a fixed reporting schedule?
Show answer
Answer: D Motion-aware scheduling can report less often while parked and more often while moving.
Retrieval practice
Recall check 6 of 6
Test Tessa says: answer from memory, then check your reasoning.
Q6A fleet geofence service needs indexed SQL point-in-polygon queries against stored geofence shapes. Why is PostGIS a suitable choice alongside timestamped tracking data?
Show answer
Answer: C PostGIS provides spatial indexes and functions such as ST_Contains for point-in-area checks.
Print reference
Answers 1 of 2
Answer key.
- A · Flexibility in component selection is crucial.
- A · This answer demonstrates understanding of: (1) power consumption trade-offs, (2) matching platform capabilities to requirements, (3) not over-engineering the solution.
- C · System integration proves the interfaces between sensing, transport, processing, and presentation instead of treating isolated component demos as a working project.
- B · LTE-M and NB-IoT suit small telemetry reports and can use PSM/eDRX to sleep between reports, saving energy.
Print reference
Answers 2 of 2
Answer key.
- D · Motion-aware scheduling can report less often while parked and more often while moving.
- C · PostGIS provides spatial indexes and functions such as ST_Contains for point-in-area checks.