Techniques for Coordination of Drone Swarms

Efficiently predicting the future trajectory of an agent in an environment using a scene-centric neural network instead of separate agent-centric networks for each agent. The scene-centric network encodes the environment and all agents using a single input centered on a fixed point. This reduces computational cost compared to separate agent-centric inputs for each agent, as the scene encoding scales linearly with the number of agents instead of quadratically.

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

Underwater robotic fish swarm for remote monitoring and virtual reality exploration of underwater environments. A floating platform communicates instructions from an operator to a submersible sinker and swarm of underwater drones to navigate and collect data in an underwater area. The collected data is transmitted back to the platform for display on a virtual reality headset worn by the operator. The swarm uses acoustic and optical communication, localization algorithms, and image compression to enable efficient underwater exploration and data transfer.

Source

A system and method for navigating fixed-wing unmanned aerial systems (UAS) in environments without or with degraded global positioning System (GPS) signals. The method uses relative motion estimation and optimization to improve local navigation and leverage occasional GPS measurements and cooperative constraints from other UAS for global positioning. The UAS estimates its motion relative to the environment using an onboard sensor fusion algorithm like an extended Kalman filter. It then optimizes a back-end pose graph representing global position by incorporating local motion estimates and occasional GPS measurements as constraints. Sharing range measurements and resetting simultaneously between UAS allows leveraging cooperative constraints. This improves accuracy compared to relying solely on local sensing in GPS-denied environments.

Source

Coordinated control of multiple distributed devices like robots using an environment manager that monitors device operations and updates environment information. The devices monitor the environment info and if it meets a condition, they execute a specific operation. This allows devices to coordinate actions based on shared environment data without central control. The environment manager updates the environment info as devices operate, allowing coordination across devices.

Source

An unmanned vehicle system that enables autonomous launch and recovery of payloads like vessels, equipment, and people in marine and submarine environments. The system uses a host vehicle like a submersible drone to carry and deploy guest vehicles like mini-subs, surface drones, and payload carriers. The host and guest vehicles can operate independently in different modes like air, surface, and sub-surface. This allows covert missions involving multiple environments and tasks. The host vehicle provides mobility, maneuverability, and launch capabilities while the guest vehicles perform specialized missions. The system enables multi-environment unmanned vehicle swarms that can be coordinated in time and space for complex missions.

Source

Using a team of drones equipped with inspection sensors to simultaneously and autonomously inspect a gas turbine engine. The drones are choreographed to move along specific paths to cover the entire engine area. If an abnormality is detected, a drone can retrace the path to confirm. If a drone takes too long, another drone is summoned to finish. This allows thorough engine inspection without engine removal or downtime.

Source

Managing flights of mobile terminals like drones that use multiple cellular networks. A central management apparatus allocates flight zones across cellular networks based on shared 3D space. This involves dividing airspace into zones identified by latitude, longitude, and altitude. The management apparatus distributes this zone allocation data to the cellular networks. Each network then uses it to coordinate flights of devices connected to their network. This ensures consistent flight management across networks when multiple devices fly in proximity.

Source

Collaborative control method for unmanned cluster systems that improves resilience against network attacks. The method takes into account parameter uncertainty and network attack impacts. It involves constructing a dynamic model, uncertainty boundary function, and adaptive robust controllers for each unmanned system. The controllers adaptively adjust parameters to compensate for uncertainty and attack effects. This allows the cluster to quickly and stably meet performance requirements in the face of parameter uncertainty and network attacks.

Source

Efficient path planning for swarms of unmanned aerial vehicles (UAVs) that balances steering the swarms towards their destinations while avoiding collisions. The approach breaks path planning into two levels: swarm-level planning for the entire swarm and individual-level planning for each UAV within a swarm. Swarm-level planning steers the swarm as a whole, avoiding conflicts with other swarms. Individual-level planning steers UAVs within a swarm towards the swarm leader and avoids conflicts between UAVs.

Source

Learning system for multi-agent applications that enables robust, scalable, and generalizable autonomous behaviors in complex environments like air-to-air engagements. The system uses population-based curriculum learning, hierarchical temporal learning, and behavior adaptation learning techniques to improve performance. It trains a diverse population of agents in sequential mini-games, learns high-level behaviors from low-level actions, and adapts behaviors to new tasks. This enables robustness, scalability, and generalizability of autonomous behaviors in multi-agent applications.

Source

Unicast air-to-everything (A2X) communications for drones that enables direct communication between drones using unicast links instead of broadcast. The drones initiate unicast links by exchanging messages over broadcast channels. The source drone sends a direct communication request (DCR) with its unique ID and application-layer ID. The target drone responds with a link establishment message. This establishes security and allows the drones to exchange unicast messages using the application-layer ID. The DCR can include proximity parameters for link setup based on drone positions.

Source

Coordinated trajectory planning for devices with multiple output components to improve control of coordinated movements. The technique involves a central component detecting trajectory errors for one component, then calculating corrected trajectories for other components to compensate. It provides look-ahead trajectories to components to reduce communication burden. The technique improves coordination efficiency and reduces network traffic in scenarios with many coordinated devices.

Source

Efficiently coordinating movement of multiple output components in a system using a centralized trajectory planning component. The method involves generating encoded output trajectories for the components, decoding them locally, and executing the decoded trajectories. This reduces communication overhead compared to streaming individual commands to each component. The centralized component calculates the trajectories based on triggers, then sends encoded versions to the local controllers for decoding and execution. This allows real-time feedback and adaptation to optimize coordination.

Source

An operation management system for efficiently and safely managing multiple vertical takeoff and landing aircraft in a shared airspace. The system uses 4D route planning to optimize flight paths considering uncertainty and external factors. It designs appropriable spaces around each aircraft that prevent collisions. The system re-plans routes during flight based on the moving and fixed appropriable spaces. This allows automated, coordinated takeoff and landing of multiple aircraft with collision avoidance and efficient use of airspace.

Source

Reducing the amount of feedback information required in coordinated wireless communication like MAP (Multi-Access Point) by having each node switch coordination modes based on local measurements instead of feeding back detailed information. Nodes measure signals from other nodes, determine if conditions warrant switching coordination modes, and transmit feedback indicating the desired mode change. This avoids flooding the air with extensive coordination feedback. The initial configuration is requested separately.

Source

A method for continuous control of multi-agent systems with complex, non-Markovian specifications using temporal equilibrium analysis and reinforcement learning. The method involves breaking down complex multi-agent tasks with LTL specifications into simpler subtasks that can be learned using reinforcement learning. The subtasks are derived through temporal equilibrium analysis, which generates abstract top-level policies. These policies are then applied to the low-level continuous control of the agents using reinforcement learning algorithms. The method improves scalability and interpretability of multi-agent systems by leveraging the strengths of both temporal equilibrium analysis and reinforcement learning.

Source

Surveillance system that optimizes placement of personnel in a surveillance area by dynamically assigning tasks to mobile devices. The system identifies areas with insufficient coverage based on device locations and notifies nearby devices to move to those areas. It repeats the process until enough devices agree to fill the gaps. By intelligently allocating resources, it can prevent under-surveilled areas from forming.

Source

Agricultural machines like tractors with integrated systems to improve work efficiency using unmanned aerial vehicles (UAVs). The machines have features like cable connections and landing stations to quickly deploy and assist UAVs during farming tasks. The UAVs can tether to the machines via cables to receive power and data. This allows the UAVs to stay longer in the air and closely monitor the machines, improving efficiency compared to separate UAVs. The UAVs can also land on the machine's landing stations to dock and recharge. This allows multiple UAVs to be connected and controlled by the machine. The UAVs can then fly in formation to cover larger areas and perform tasks like crop spraying or scouting. The integrated UAV systems avoid the issues of separate UAVs like limited flight time, restricted monitoring range, and collisions.

Source

A control system for coordinating multiple robots that reduces communication bandwidth and storage requirements compared to traditional centralized multi-robot systems. The system uses a central control apparatus to manage the robots instead of having each robot communicate with all others. The control apparatus assigns tasks to individual robots based on their observed environments. This decentralized task allocation reduces the need for inter-robot communication and avoids propagating all sensor data. The control apparatus can also store and process the raw sensor data, further reducing robot memory requirements.

Source