GPS-Denied VIO SLAM Mapping

GPS-denied environments such as forests, indoor corridors, urban canyons, tunnels, or disaster zones, autonomous drones cannot rely on satellite positioning for navigation. To solve this problem, the proposed GPS-denied VIO SLAM mapping system combines Visual–Inertial Odometry (VIO), onboard camera data, IMU measurements, and CNN-based scene interpretation to estimate the drone’s motion and build a local map in real time. The drone uses its onboard camera to observe visual features in the environment, while the IMU provides acceleration and rotation measurements. By fusing these data sources, the system estimates the drone trajectory even when GPS is unavailable. The VIO module tracks visual features across image frames, estimates relative motion, and continuously updates the drone's pose. At the same time, the CNN-based perception layer helps identify free space, vegetation, obstacles, and navigable regions. The mapping interface shows the drone’s live forward navigation view, detected visual features, free-space score, vegetation percentage, and the estimated trajectory on a 2D local map. The blue path represents the reconstructed drone trajectory, while the surrounding point cloud indicates detected environmental structure. This allows the UAV to maintain situational awareness and continue navigation in complex terrain where GPS signals are blocked or unreliable.

https://video.wixstatic.com/video/300297_438a8d6caad0478cb35ba6c583cdc639/720p/mp4/file.mp4

During flight, the interface displays the live onboard camera view, navigation direction, feature points, free-space score, vegetation percentage, and the estimated 2D trajectory. The blue line represents the drone’s reconstructed path, while the surrounding point cloud shows detected environmental structure. This enables the UAV to maintain stable navigation, avoid dense vegetation, and continue mapping even when GPS signals are unavailable.

This approach is especially useful for search-and-rescue missions, forest inspection, disaster response, and autonomous exploration, where drones must operate safely without external positioning infrastructure. By integrating VIO, CNN perception, and SLAM-based mapping, the system provides a resilient navigation framework for autonomous UAV operation in GPS-denied scenarios.

Cyber–Physical Attacks on Drones and Mitigation Strategies

Drones are no longer only flying cameras. They are cyber–physical systems that combine embedded computers, wireless communication, sensors, actuators, navigation algorithms, and autonomous decision-making. This integration makes unmanned aerial vehicles useful for inspection, mapping, search and rescue, agriculture, defense, and logistics. However, the same connectivity and autonomy that make drones powerful also expose them to cyber–physical attacks. A cyber–physical attack targets both the digital and physical behavior of the drone. Instead of only stealing data or disrupting software, the attacker may influence how the drone moves, where it navigates, what it senses, or how it communicates with other agents. In the worst case, this can cause mission failure, collision, loss of control, or unsafe behavior in real environments.

One common attack surface is the communication link between the drone and the ground control station. If the command channel is not protected, an attacker may inject false commands, interrupt telemetry, replay old packets, or jam the wireless signal. For swarm-drone systems, drone-to-drone communication is also critical. A compromised link can affect formation control, leader–follower coordination, shared mapping, and collaborative decision-making.

Another important threat is navigation spoofing. Drones that depend on GNSS/GPS can be misled by fake satellite signals or denied access through jamming. In GPS-denied scenarios, attackers may also target visual, inertial, LiDAR, or radar-based navigation. For example, adversarial visual patterns, sensor saturation, or false obstacle information can degrade perception and cause incorrect path planning. Sensor attacks are especially dangerous because autonomous drones depend on real-time perception. A LiDAR sensor can be affected by reflective surfaces, interference, or spoofed distance measurements. Cameras can be affected by lighting manipulation, adversarial markers, smoke, fog, or occlusion. IMU and magnetometer readings can also be disturbed, leading to drift in attitude or position estimation.

Mitigation requires a multi-layer defense strategy. First, communication channels should use authentication, encryption, packet integrity checks, and anti-replay mechanisms. Every command and telemetry packet should be verified before it is accepted by the drone. For swarm systems, leader-to-follower messages should include sequence numbers, timestamps, source and destination identifiers, RSSI, latency monitoring, and packet delivery ratio estimation. Sensor fusion can improve robustness by combining GNSS, visual-inertial odometry, LiDAR, radar, barometer, magnetometer, and onboard mapping. If one sensor becomes unreliable, the system can switch to a degraded but safe navigation mode. For GPS-denied missions, visual-inertial odometry, SLAM, LiDAR mapping, and local obstacle avoidance are key tools.

Anomaly detection should be integrated into the control loop. The drone should continuously monitor unexpected changes in position, velocity, heading, communication quality, sensor readings, and actuator behavior. If the system detects abnormal telemetry, packet loss, spoofing symptoms, or inconsistent sensor fusion results, it can activate fail-safe behaviors such as slowing down, hovering, returning to a safe waypoint, landing, or switching to manual control.

Fourth, resilient control algorithms are needed. Controllers should be designed to tolerate disturbances, packet loss, delayed commands, and sensor uncertainty. Techniques such as robust control, adaptive control, fault-tolerant control, and learning-based decision modules can help the drone maintain stability under degraded conditions. In swarm navigation, followers should be able to maintain formation using the last trusted leader state while avoiding unsafe behavior when communication becomes stale.

Cybersecurity must be considered during the design stage, not added only after deployment. Secure firmware updates, hardware root of trust, protected boot, access control, logging, intrusion detection, and simulation-based attack testing should be part of the drone development workflow. Digital twins and simulators are useful for testing cyber–physical attacks before real-world deployment.

In conclusion, drones must be protected as complete cyber–physical systems. Securing only the software or only the wireless link is not enough. A robust drone architecture should combine secure communication, sensor fusion, anomaly detection, resilient control, and fail-safe mission logic. As drones become more autonomous and collaborative, cyber–physical security will be essential for safe and reliable operation in real-world environments.

---

AeroSwarm: A GPS-Denied Multi-UAV Simulation Framework

AeroSwarm is a simulation framework designed to study collaborative drone navigation in GPS-denied environments. The platform models a swarm composed of one leader drone and two follower drones operating in a complex scenario with obstacles, limited visibility, and communication constraints.

https://video.wixstatic.com/video/300297_12c8844e8030442f81f102cc2d5b6580/720p/mp4/file.mp4

The leader drone is equipped with a 360° LiDAR sensor that continuously scans the environment, detects obstacles, and supports real-time navigation. During the mission, the leader transfers its position, waypoint information, velocity, LiDAR status, and navigation state to the follower drones through drone-to-drone communication links. This allows the swarm to maintain formation and continue coordinated navigation even when GPS is unavailable. The simulator includes multiple dashboard tabs for monitoring the mission. The Navigation + Mapping tab shows the camera views, HUD information, and swarm scenario map. The LiDAR + COMM overlay tab visualizes the LiDAR rays, obstacle detection, and communication links. The Sniffer tab monitors the packet-level communication between the leader and follower drones, including RSSI, latency, packet delivery ratio, RX/DROP state, and LOS/OCC channel condition. AeroSwarm is useful for testing autonomous navigation strategies, communication-aware formation control, LiDAR-based perception, and swarm coordination under GPS-denied conditions. It provides a flexible research platform for developing robust multi-UAV systems for search and rescue, inspection, defense, and operations in indoor, urban, forest, or degraded environments.

---

https://github.com/1Px-Vision/AeroSwarm