Fall Test Plan
Spring Validation Demonstrations
Spring Validation Demonstration 1
| Description | UAV Apriltag following and obstacle avoidance in Gazebo Simulation |
| Location | Virtual |
| Equipment | Master Computer |
| Setup | Place Apriltag about 10m away from the UAV; Place box obstacles between UAV and Apriltag |
| Procedure | 1. UAV was given the GPS coordinates of the Apriltag 2. UAV takes off and approach the Apriltag while avoiding obstacles 3. When the UAV is within 1m radius of the Apriltag, switch to Apriltag tracking 4. UAV land on the Apriltag |
| Criteria | 1. UAV avoids all obstacles (M.P. 2) 2. UAV lands on Apriltags with error < 0.15m |
Spring Validation Demonstration 2
| Description | UGV Decision Making and Automatic Navigation in Gazebo Simulation |
| Location | Virtual |
| Equipment | Master Computer |
| Setup | A map built by static obstacles, which is known to the robot; a map with added dynamic obstacles; |
| Procedure | 1. Initialize UGV within a map of static + dynamic obstacles 2. UGV receives a landing command and automatically navigates towards the computed optimal landing point 3. Near the landing point, UGV automatically explores updated dynamic obstacles, and decide whether it is feasible for landing 4. If not feasible, the resume step 2 as UGV navigates towards another landing point candidate |
| Criteria | 1. No collision while UGV is travelling (M.P.2) 2. Reach the target state within 60s (M.P.1) 3. Reach the target state with (x, y) error less than (0.3m, 0.3m) (M.P.3 & M.P.4) |
Spring Validation Demonstration 3
| Description | UAV field test on GPS waypoints following (Depend on COVID 19 Situation) |
| Location | Boyce Park |
| Equipment | UAV, tape, measuring tools |
| Setup | Mark the initial position of the UAV |
| Procedure | 1. UAV takes off 2. UAV follows waypoints that forms a regular hexagon with side length 20m 3. UAV finishes loop and lands |
| Criteria | 1. UAV finished the hexagon loop and lands within 1m of the starting point (M.P. 9) |
Spring Validation Demonstration 4
| Description | Manual Landing onto Apriltag Platform (Depend on COVID 19 Situation) |
| Location | Boyce Park |
| Equipment | DJI M100 UAV, Landing Module |
| Setup | Fix the landing module onto the ground; place the UAV onto the landing module |
| Procedure | 1. Power on the UAV and establish remote connection 2. Switch off electromagnets, maneuver the UAV to takeoff 3. Fly out the UAV for 100m, and use GPS to return to home 4. Maneuver the UAV to approach to the landing platform, and adjust its pose to align with receptors 5. Switch on electromagnets, and UAV lands into receptors |
| Criteria | 1. The landing process success rate is over 60% (M.P.1) 2. The GPS return-to-home will generate an error of less than 1m (M.P.9) |
Fall Validation Demonstration
Fall Validation Demonstration
| Description | Collaboration path planning and landing between the UAV and the UGV |
| Location | CMU The Mall |
| Equipment | DJI M100 UAV, Jackal UGV, cardboard boxes |
| Setup | 1. Place the UAV and the UGV 100m aprt in an open area 2. Place piled boxes between the UGV and the UAV |
| Procedure | 1. Initialize UAV and UGV in a mapped area with the knowledge of static obstacles 2. With proper control input into the electromagnets, the UAV takeoff from UGV landing platform 3. The UAV flies off to reach certain predefined waypoints, with battery status (could be simulated) taken into account 4. As the UAV completes its goal and battery is low, the UAV sends a signal to ROS master, and the master will dispatch a UGV for landing mission 5. The UGV computes optimal landing point candidates, sends a message to the UAV, and automatically navigate towards one of the candidates 6. The UAV will fly towards the chosen landing area, using its sensing mechanism, to evaluate initial landing feasibility 7. If 6 is feasible, the UGV will continue navigating to the targeted landing area; it will use the perception module to validate landing feasibility 8. If 7 is feasible, the UAV will start tracking Apriltag to compute transformation and plan optimal landing path 9. The UAV will execute an optimal landing path with the computed frame transformation; it will approach the landing platform in a vertically descending manner, and adjust its position/orientation to fit its feet into the cone-shape receptors 10. As the distance to landing platform is close enough, apply proper control input to the electromagnet, the UAV will be absorbed into the landing receptor, and the landing process completes |
| Criteria | 1. UAV have a safe and steady land on UGV without tipping over (M.P. 3) 2. UAV land on UGV within 60 seconds (M.P. 1) 3. UAV and UGV avoid all obstacles (M.P. 2) 4. Neither UAV or UGV ran out of battery (M.P.7) |