AUV INTEGRATION
UNDERWATER VEHICLES IN LISA 26

Operational Concept — Underwater Drone

An AUV launches from shore or from a boat, descends below the surface, and follows a pre-programmed search pattern using its inertial navigation system and Doppler velocity log. It scans the seabed and water column with side-scan sonar, recording everything to onboard storage. After completing its mission (typically 2-8 hours for small tactical AUVs), it surfaces and returns to a recovery point. The operator extracts the data via USB or WiFi and debriefs it through Lisa 26's offline pipeline — the same process used for fiber-optic FPV and RF-silent ISR drones.

Baltic Sea Context

Sweden's coastline and the Baltic Sea are critical for naval defense. AUVs can map harbor approaches, detect mines or underwater obstacles, monitor submarine activity, and survey infrastructure (pipelines, cables). The shallow, cold Baltic (average depth 55m, winter temperatures near 0°C) presents specific challenges: limited visibility, strong currents in straits, and ice cover in winter.

Dead Reckoning — Navigation Without External Signals

Underwater, radio waves attenuate within meters in salt water. GPS signals do not penetrate the surface. Acoustic positioning systems (USBL, LBL) require pre-deployed transponder arrays. The AUV must navigate independently using dead reckoning: integrating acceleration from the IMU (Inertial Measurement Unit) to estimate velocity, then integrating velocity to estimate position. The DVL (Doppler Velocity Log) improves this by measuring speed relative to the seabed using four acoustic beams pointed downward at different angles.

DVL accuracy depends on proximity to the seabed. At depths under 200 meters (typical for Baltic Sea operations at 40-100 meter depths), DVL provides velocity accuracy of 0.1 percent of distance traveled — an AUV traveling 1 km accumulates 1 meter of position error from DVL alone. Combined with IMU drift: total position uncertainty of approximately 50 meters per hour of transit. For a 30-minute mission covering 2 km: position error of 25-50 meters at the end of the run. Sufficient for area reconnaissance but not for precise target localization.

Operational Concept — Covert Reconnaissance

The AUV launches from shore or from a small boat at a known GPS position. It submerges and follows a pre-programmed route defined as a sequence of waypoints in LOCAL_NED coordinates relative to the launch point. The route covers the area of interest: a harbor entrance, a suspected mine corridor, an underwater cable route, or a coastal defense installation. The AUV records sonar imagery, water temperature profiles, bottom composition, and optical imagery (if equipped with a camera and lighting) to its encrypted SD card.

Upon completion, the AUV surfaces at a recovery point (which may differ from the launch point for operational security). The operator retrieves the SD card and runs the Lisa 26 debrief script — the same one used for expendable ISR drones. Detections are imported into the brigade COP retroactively with timestamps. The data is 30-60 minutes old — not real-time intelligence. But for mapping harbor approaches, identifying underwater obstacles, and confirming or denying suspected mine placements, delayed intelligence is far better than no intelligence. The alternative is sending combat swimmers to gather the same data at enormous personal risk.

Lisa 26 Debrief Pipeline for AUV

AUV debrief follows the identical pipeline as fiber-optic FPV and RF-silent ISR (see §6.5 Offline Debrief). After surfacing, connect AUV data storage via waterproof USB connector. Run lisa26-extract --platform=auv --input /dev/sdb1. The tool recognizes side-scan sonar format (XTF or JSF) and converts to georeferenced imagery. Sonar contacts are classified as: mine-like object, wreck, obstruction, or seabed feature. Operator confirms or rejects. Confirmed contacts appear on COP with underwater symbology.

← Del av Ekf3 Sensor Fusion

External source: UUV – Wikipedia

Implementation

# AUV Dead Reckoning Navigation — No Radio Underwater
# pip install numpy
import numpy as np
import time

class AUVDeadReckoning:
    def __init__(self, start_lat, start_lon, depth_m):
        self.lat = start_lat
        self.lon = start_lon
        self.depth = depth_m
        self.heading_deg = 0
        self.speed_ms = 0
        self.log = []
    
    def update(self, imu_heading_deg, dvl_speed_ms, dt_s):
        """Update position from IMU heading + DVL speed."""
        self.heading_deg = imu_heading_deg
        self.speed_ms = dvl_speed_ms
        
        dx = self.speed_ms * np.cos(np.radians(self.heading_deg)) * dt_s
        dy = self.speed_ms * np.sin(np.radians(self.heading_deg)) * dt_s
        
        self.lat += dy / 111320  # meters to degrees
        self.lon += dx / (111320 * np.cos(np.radians(self.lat)))
        
        self.log.append({
            "time": time.time(),
            "lat": self.lat, "lon": self.lon,
            "depth": self.depth, "heading": self.heading_deg
        })
    
    def save_to_sd(self, path="/data/auv_mission.jsonl"):
        """Save mission log to SD card for offline debrief."""
        import json
        with open(path, "w") as f:
            for entry in self.log:
                f.write(json.dumps(entry) + "\n")
        return len(self.log)

# AUV surfaces → SD card retrieved → Lisa 26 debrief imports data
# Zero RF emission entire mission. Undetectable underwater.

Related Chapters

Sources

See the categorized source sections earlier on this page for specific citations supporting each claim. Cross-referenced technical baselines: ArduPilot developer documentation; ExpressLRS hardware documentation; NATO STANAG 4609 Ed. 4 (motion imagery metadata), 4671 (UAV airworthiness), 2022 (intelligence evaluation); Watling & Reynolds, "Meatgrinder: Russian Tactics in the Second Year of Its Invasion of Ukraine", RUSI (2023); ISW daily campaign assessments at understandingwar.org (archive). FSG-A has no own operational experience.