Spring videos:
Full system demo:
Signature Detection Result in RGB Video :
Signature Detection Result in Thermal Video :
Bright-colored object detection:
Hot objects detection using IR imagery:
Spring Validation Experiment
During the SVE, our system’s performance was validated as per the designed test sequences.
All our subsystems were successful on all of their respective performance measures.
This section summarizes the overall system performance during the SVE and SVE Encore. For detailed analysis, refer to the analysis done for each of the subsystems.
| Procedure | Procedure | Verification Criteria | SVE Performance | SVE Encore Performance |
|---|---|---|---|---|
| 1. | Place seven signatures specified above, at various locations in a 50m x 50m area.waypoints | |||
| 2. | Place UAV on the ground, turn it and the payload power ON. Create and launch a search mission through the Mobile app. | |||
| 3. | UAV sweeps the area (50m x 50m) collecting sensor data. | |||
| 4. | Transfer data from the payload to the laptop. | |||
| 5. | Run integrated human detection software to detect human signatures. | The system should be able to detect at least 5/7 human signatures planted | 6/7 signatures detected. The system was able to detect 1 out of 2 humans and all the other signatures | 7/7 signatures detected. The system was able to detect all the signatures |
| 6. | Run software to report GPS coordinates of the probable rescue locations. | The system should be able to report GPS coordinates of the rescue locations with a margin of error of +-8m | We got likely locations within +/- 5m of all signatures just except for the human. Human was reported to be 8.65 m away from actual | We got likely locations within +/- 3m of all the signatures |
| 7. | Select one location out of the reported rescue locations for the package drop | |||
| 8. | Detach the microphone from the payload, attach the rescue package. Launch rescue package drop mission through the Mobile app. | |||
| 9. | UAV flies to the provided GPS location, lands, and autonomously drops the package. | Rescue package should be dropped with distance less than 8m from the signature. (assuming the signature has not moved) | Autonomous package drop did not work, but the drone landed 2 m away while testing for a bright mattress, and 9 m away for the human | Autonomously dropped the package 2.3 m away from a human location |
| 10. | UAV flies back to the home location and lands. | Total time < 25 minutes | Total time > 25 minutes | Total time < 25 minutes |
Fall Validation Experiment
During the FVE, our system’s performance was validated as per the designed test sequences.
All our subsystems were successful on all of their respective performance measures.
Following table details out the requirements for the various test steps for each of our subsystems and our subsystems’ performance on them.
| Steps | Requirements | Subsystem | Performance |
|---|---|---|---|
| A.1. | Autonomous Flight System | iOS App to provide input GPS | |
| A.2. | Accuracy in reaching desired height (+-1m tolerance) | Autonomous Flight System | Height provided = 5M. Min height = 4.8M, Max height = 5.3M |
| A.3. | Accuracy in reaching the waypoints (+-5m tolerance) | Autonomous Flight System | Accuracy : <3M from each waypoint |
| A.4. | Accuracy in reaching the starting location (+-5m tolerance) | Autonomous Flight System | Back to start location : <3M |
| B.1. | Ability to detect at least 60% of valid human signatures in images | Signature Detection | Overall accuracy for 40 test images: 65.1% |
| C.1. | Should be able to hold package of weight 100g, and size 10cmx10cm | SDPD (Package Drop Mechanism) | Successful (package weight ~160g) |
| C.2. | Should not lose grip of the package for drone velocities under 2 m/s | SDPD (Package Drop Mechanism) | Successful |
| C.3. | Manually demonstrate the mechanism’s ability to release the package | SDPD (Package Drop Mechanism) | Successful (demonstrated switching through commands sent from phone) |