Emergency Drone Landing Systems

Parachute landing system for UAVs that enables controlled deployment of the parachute while maintaining precise control over the landing process. The system integrates with the UAV's flight control system to determine landing conditions and execute the parachute deployment sequence, including stopping the motor, deploying the parachute, and stabilizing it for a predetermined period. This controlled deployment enables precise landing parameters while maintaining flight control authority.

Source

This paper presents a strategy for collision detection and recovery control of drones using an onboard Inertial Measurement Unit (IMU). The algorithm compares the expected response drone with measurements from IMU to identify characterize collisions. controller implements gain scheduling approach, adjusting its parameters based on characteristics drones attitude. Simulations were conducted compare proposed popular method fixed thresholds, simulation results showed that approach outperformed existing in terms accuracy. Furthermore, approaches tested physical experiments custom-built drone. experimental confirmed was able distinguish between actual collisions aggressive flight maneuvers, can recover within 0.8 s.

Source

Method for safely terminating navigation of an unmanned aerial vehicle (UAV) that ensures safe landing even when the UAV deviates from its planned route. The method generates a precise navigation plan for the UAV, which includes a start point, end point, and a virtual tunnel connecting these two points. The UAV executes the navigation plan by following the planned route, and when it reaches the end point, it terminates its flight path by disconnecting power to the propulsion system and deploying a failsafe, such as parachutes or airbags. This approach ensures safe landing even when the UAV deviates from its planned route, providing a failsafe for civilian drone operations.

Source

Autonomous landing of drones in complex environments using adaptive path planning and deep reinforcement learning. The method selects between local and global path optimization based on perception range. For local optimization, drones plan paths around nearby obstacles. For global optimization, they use perceived frontiers. This improves efficiency by avoiding redundant planning. For landing, a neural network learns to control the drone using reward functions. This increases neural network update efficiency compared to traditional methods.

Source

Flight control method to reduce occurrence of abnormal situations like flipping over after collision or drifting pitch during aircraft flight. The method involves detecting collisions using disturbance monitoring or model calculations, then in response reducing motor speeds to lower altitude and adjusting attitude to normal. This prevents getting stuck on obstacles after collision and prevents pitch drift by lowering altitude.

Source

Safe landing point search method for unmanned aerial vehicles (UAVs) on unfamiliar terrain using terrain maps and contour lines. The method involves generating contour lines based on a minimum height and interval from a terrain map. Then, it searches for largest empty circles (LECs) with a minimum radius in the contour map. The UAV is provided the LEC with the largest radius as a safe landing spot. This leverages contour lines to find flattest areas without obstructions for UAV landing.

Source

Optimizing communication in swarms of drones to enable efficient and reliable control of large numbers of drones. The optimization involves clustering the drones into groups with a master drone that communicates with a central server, and slave drones that relay messages from the master. This reduces the number of required communication channels compared to each drone directly connecting. Clustering also allows faster area coverage, obstacle avoidance, and resource sharing. If a master fails, another slave can be promoted. This enables robust swarm operation by minimizing communication breakdowns.

Source

Enabling high-priority message broadcasts to autonomous devices like drones and self-driving cars in wireless networks like 5G. The broadcasts contain commands for the autonomous devices to execute in response to emergency alerts. The devices receive a notification of an upcoming high-priority message, then the actual message with commands. This allows autonomous devices to respond to emergencies even if no human is present. The messages can also request responses from devices.

Source

Detecting and responding to flight-restricted regions for unmanned aerial vehicles (UAVs) to permit automated flight control in response to detected proximity to restricted areas. The method involves calculating distances between the UAV and flight-restricted regions using location data. If the distance falls within certain thresholds, flight response measures like landing or preventing takeoff are initiated. This provides automated response to restricted areas with graded actions based on proximity. It also uses location systems to accurately detect restricted areas.

Source

Unmanned Aerial Vehicles (UAVs), or drones, have become increasingly important in various fields, including aerial photography, delivery services, and surveillance. However, as drone usage has expanded, so the risks related to system failures accidents. Such can result expensive damages even threaten public safety. A promising way address these is by implementing a recovery that uses parachute mechanism. This paper discusses design, development, implementation of parachute-based for drones. It looks into essential components system, such selection, deployment methods, material choices, performance testing. The goal improve safety operations ensuring controlled descents during emergencies, thus minimizing risk crash-related damage.

Source

Aerial network architecture using hybrid communication links like FSO and RF that allows simultaneous transmission and failover protection. The network nodes like drones have overlay networks with FSO transmitters/receivers and RF transmitters/receivers. A processor modulates data for both links and balances traffic preemptively based on mission planning and link conditions. This reduces congestion and packet loss during link failures. The processor also uses packet erasure coding to recover from spurious FSO link loss without needing error prediction. The hybrid links and techniques provide resilient and efficient networking for aerial platforms.

Source

Autonomous transition system for electric vertical takeoff and landing (eVTOL) aircraft that allows smooth and safe transition between vertical and horizontal flight. The system has a flight controller that takes over control after takeoff to transition the aircraft from vertical to horizontal flight. It uses a preplanned trajectory to generate torque in the vertical lift component and adjust the attitude. Then it engages the horizontal thrust component and modulates the lift component. The pilot can override the autonomous mode at any time.

Source

Automatically selecting alternate destinations for robots in response to contingency events during travel, to improve robot autonomy and reliability. The method involves detecting a contingency event like a failure or obstacle, determining the robot's current position, accessing alternate destinations associated with the route, selecting the best alternate based on travel time, and outputting the new destination for navigation. This allows the robot to safely continue the mission instead of being stranded.

Source

Redundant electric VTOL aircraft that can maintain safe flight and landing even if one motor fails. The aircraft has multiple vertical takeoff/landing motors and horizontal flight motors. If a vertical motor fails during takeoff or landing, the remaining vertical motors can still provide lift. If a horizontal motor fails during cruise, the other horizontal motors can continue propulsion. The aircraft has separate high-capacity power sources for horizontal flight and high-power backup sources for vertical takeoff/landing. This allows redundancy and continued safe flight without having to double all systems.

Source

Obstacle avoidance for unmanned vehicles using monocular cameras and deep learning. The technique involves selecting a sub-image from the main camera feed for obstacle detection based on field of view and image size. This sub-image is processed with a low accuracy depth estimation model to compute an average depth. If the average depth is below a threshold, an obstacle is detected and avoidance maneuvers are initiated. This allows precise obstacle avoidance using monocular cameras instead of expensive sensors like LiDAR.

Source

Intercepting and controlling drones without damaging them by redirecting them to land using spoofing signals. The method involves detecting unauthorized drones near restricted areas, deploying police drones to intercept them, and transmitting signals to program new flight paths to land. The redirecting signals include spoofing GPS signals to simulate satellite constellations and jamming signals to disable the target drone's controls. The police drones position themselves close to the targets and then transmit the redirect signals to safely guide the intruders to designated landing areas.

Source

Intelligent method for protecting the battery of an electric unmanned aerial vehicle (UAV) in real-time by calculating the needed electricity for safety actions based on the current position. If the UAV's remaining battery charge is less than the calculated safety amount, it triggers actions like landing or returning. This protects against low battery accidents. The safety amount is dynamically calculated based on the UAV's position to optimize protection versus flight range. It improves UAV battery utilization compared to fixed low voltage alarms.

Source

Landing facility for drones that enables safe landings even in strong winds. The facility has a designated landing area with a windbreak part surrounding it. The windbreak part is tall enough to shield the landing area from winds but not so tall that it generates vortexes behind it. This prevents instability during takeoff and landing. A second area beyond the windbreak is where the drone descends to a certain altitude before landing. This allows the drone to transition from windy conditions to calm air near the landing zone.

Source

Safe landing assistance system for aerial vehicles like drones that provides intuitive displays and alerts to help pilots land accurately in constrained urban environments. The system shows the vehicle's proximity to the landing zone center and time/altitude to touchdown. If the vehicle deviates or runs out of time, it alerts to abort. This allows pilots to adjust orientation and descend until an abort point. After that, it's too late to land safely, so the system computes flight controls to modify the landing.

Source

Failover handling in autonomous vehicles to allow them to continue operating when primary planning systems fail. The method involves having a fallback path planner that calculates paths independently of the primary planner. A path multiplexer selects between primary and fallback paths based on normal vs degraded operating conditions. If the primary planner enters a degraded state, the fallback path is used instead. This allows the vehicle to continue moving without relying on remote assistance when primary planning fails.

Source