UC02.2: Indoor Drone Exploration and Victim Detection (USAR)

Last modified by Tjalling Haije on 2025/09/15 09:25

Indoor Drone Exploration and Victim Detection – USAR

Objective	OB01: Increase effectiveness through exploration, sensing and communication means OB04: Improve safety by providing overviews and warnings of environment conditions Use an indoor drone to explore buildings before entry, locate victims, and update the C3I map for safer and faster intervention.
TDP	TDP1, TDP2, TDP4, TDP5
IDP	IDP1
Actors	Drone Operator, Analyst, Team Leader, USAR Rescuers (described at 4. Personas & Problem Scenarios and Direct Stakeholders)
Pre-Condition	Structure is unsafe for entry; drone is available.
Post-Condition	Victim locations marked; team briefed.
Status	Validated during and refined after CFT 1, CFT 2, CFT 3

Action Sequence

Deployment and Startup
a. Team leader identifies structure as unsafe and requests drone deployment.
b. Drone operator sets up and tests the drone, connects it to the C3I system.
c. Analyst prepares the interface to process and annotate live data.
Initial Entry and Autonomous Navigation
a. Drone autonomously navigates through the structure using onboard sensors.
b. Drone detects victims, hazards, and structural features.
c. Data is transmitted to the analyst in real time.
Live Monitoring and Victim Identification
a. Analyst verifies detections and flags victims or hazards.
b. If needed, operator switches to manual control for detailed exploration.
c. Confirmed findings are tagged on the C3I map.
Data Marking and Coordination
a. Analyst shares tagged map and video with team leader.
b. Team leader updates the plan and briefs USAR team.
c. Drone finishes scan and returns or continues overwatch mode.

Claims (title)	Function	Effect(s)	Action Sequence Step(s)
CL1 Video stream improves pre-entry SSA	Live video	EFXX: Situation Awareness is improved. → Measured with SAGAT method and questionnaire. (Oude effect van Mark tussen haakjes steeds: Team enters with real-time building and victim data)	4a
CL2 Map improves pre-entry SSA	C3I mapping, LIDAR	Same as above	3c, 4a
CL3 Map improves responder safety	Drone maps hazard zones and structural integrity	EFXX: First Responder 5. Safety is improved → Measured with near incident analysis. Note: is this function "victim detection" talking about that function of the human + drone together as one system. Or only of the technology, i.e. the drone? (Minimizes exposure to unsafe zones)	2b, 3a, 3b, 3c, 4a
CL4 Autonomous victim detection improves team performance and safety	Autonomous navigation, Object detection (Thermal imaging?)	Victims are detected more quickly, accurately and robustly than when the robot is directly controlled by a human. → Measured with time until area mapped + victims correctly localized. Note: Verwachten we hier echt dat dit sneller gaat dan een mens die het remotely operate? Of een mens die naar binnen gaat en het zelf doorzoekt? Autonoom is mogelijk wel robuuster bij e.g. verbinding verlies. EFXX: Task performance increased. → Measured with time until area mapped + victims correctly localized. (vs. e.g. human doing all victim detection on their own using camera stream). EFXX: Mental workload of drone operator and FR is low → measured with NASA TLX questionnaire. Note: Wat doen de drone operator en FR nu in de tussentijd? Wachten of meekijken? Is dat beter dan voorheen? (Speeds up rescue prioritization)	2a, 2b, 3a
CL5 Analyst involvement improves SA and trust calibration	Analyst verifies and adds missing detections and victims on camera live video stream	EFXX: Trust (or reliance?) in SA is good → Measured with ??. Note: meeste measures meten het vertrouwen in de robot, maar hier specifiek geïnteresseerd in het vertrouwen in het eindersultaat?	3a
CL6 Adaptive autonomy ensures acceptable workload for drone operator and analyst	Adaptive autonomy	EFXX: Mental Workload of drone operator acceptable → Measured with NASA TLX questionnaire.	3a, 3b

Mogelijke andere claims:

Analyst verification zorgt voor geen responsibility gap