Unmanned aerial vehicle pod data acquisition system based on robot operating system

Abstract

The invention discloses an unmanned aerial vehicle pod data acquisition system based on a robot operating system, and designs a pod architecture for field data acquisition, and integrates a microwave wake-up module, a microstrip patch antenna array and a low-power-consumption circuit to ensure stable and continuous communication. Through a flight control algorithm developed based on ROS, a data acquisition path is dynamically optimized under the consideration of factors such as camera distribution, pod residual electric quantity and wind field data. A communication protocol layer is innovated, a two-stage wake-up mechanism is adopted, the first stage broadcasts a low-power-consumption Bluetooth BLE signal, the second stage adopts microwave directional beam scanning, the effective wake-up distance is increased, and the false wake-up rate and the comprehensive power consumption are reduced. According to the invention, the coverage rate of single field data acquisition and the data transmission rate can be improved, a stable and high-precision automatic technical scheme is provided for the absence of a special module for field data acquisition in the unmanned aerial vehicle pod at present, and the intelligent development of an ecological protection technology and the deepening of the application of the unmanned aerial vehicle in a special operation scene are promoted.

Description

Technical Field

[0001] This invention relates to the fields of drone applications and communication technology for field monitoring equipment, specifically to a drone pod system that integrates a robot operating system, microwave communication, and large-capacity storage for automatically collecting monitoring data from field infrared cameras. Background Technology

[0002] Currently, field monitoring data collection mainly relies on manual on-site collection of equipment: staff must frequently venture into the field to manually retrieve image data from infrared cameras. This method is difficult to cover a large number of devices in a single operation, resulting in low monitoring efficiency and safety risks in harsh environments or complex terrain. Some attempts to use 4G network transmission solutions have failed completely in remote protected areas without public network signals, leading to data transmission delays or even loss, making it difficult to meet real-time monitoring needs.

[0003] Existing general-purpose drone pods are primarily designed for aerial photography scenarios and lack dedicated functional modules for field data acquisition. Their standard communication interfaces cannot establish efficient data interaction with infrared cameras, resulting in low efficiency and success rate in waking up devices, poor anti-interference capabilities, and low transmission rates. Furthermore, their path planning algorithms for acquiring data from multiple field cameras under multi-source conditions are inadequate. Simultaneously, the high power consumption of the pod's continuous communication module or other functional modules significantly depletes the drone's power resources, compressing operational time and making it difficult to support large-scale monitoring tasks.

[0004] To address the aforementioned problems, this invention proposes a method for acquiring field monitoring data using an infrared camera based on a mobile device. This method solves the technical bottlenecks in UAV applications and field data acquisition through a three-dimensional innovation of dedicated hardware architecture, flight control algorithms, and hierarchical communication protocols: the dedicated pod hardware integrates a microwave wake-up module, using a microstrip patch antenna array to achieve directional wake-up within 150 meters using a 30° beam in the 5.8GHz band. Furthermore, a two-level wake-up mechanism is employed at the communication protocol layer: BLE broadcast coarse scanning and microwave directional beam transmission of a 16-bit device ID Gold code sequence for fine wake-up, with hardware and software integration ensuring stable device wake-up. A dynamic bias circuit is designed to reduce power consumption, and combined with an independent power supply system, the flight endurance is improved by 40%; the ROS intelligent control system develops an adaptive PID controller and designs a multi-objective genetic algorithm for path planning. Path points are selected based on constraints such as battery level, communication blind spots, and no-fly zones, and an energy-sensing crossover operator is added to ensure that the selection of crossover points is positively correlated with the current battery level. The above invention provides a high-coverage, high-stability, and high-efficiency solution for field data acquisition.

[0005] In summary, this invention solves the problems of low efficiency of manual data collection, lack of general-purpose pod data collection function, and insufficient battery life in field monitoring by integrating a directional microwave wake-up module and reconstructing a low-power circuit, and introducing the ROS flight control algorithm and hierarchical wake-up protocol. Summary of the Invention

[0006] This invention is a data acquisition system for unmanned aerial vehicles (UAVs) pods based on a robot operating system, addressing the problems of poor specialization, low communication efficiency, and short endurance in existing UAV systems for field data acquisition. The method innovatively designs a dedicated pod hardware architecture, integrating a high-frequency directional microwave communication module and a large-capacity storage unit. Through dynamic voltage regulation technology and pulsed transmission signals, it reduces standby and overall power consumption. Combined with an intelligent flight control algorithm developed based on ROS, it achieves automated and efficient data acquisition from a network of infrared cameras in the field. Regarding communication protocols, the system employs a BLE signal and Gold-encoded sequence verification mechanism to achieve precise device wake-up within a 150-meter range. It innovatively designs a multi-factor constrained path planning algorithm to simultaneously optimize three key indicators: flight path length, energy consumption, and communication quality. Furthermore, it establishes a hierarchical communication protocol stack, improving data transmission efficiency and increasing single-operation coverage through BLE-WiFi6 dual-mode collaborative transmission and adaptive time slot allocation mechanisms. This system achieves specialized, intelligent, and efficient integration of UAV pods in field data acquisition scenarios, providing a complete technical solution for the field of ecological protection.

[0007] The technical solution of this invention is: a data acquisition system for a drone pod based on a robot operating system, comprising the following modules and algorithms:

[0008] (1) Microwave wake-up module:

[0009] This module, as one of the core components of this invention, is more stable and accurate than traditional solutions. It supports precise wake-up within 150 meters through directional beamforming and reduces power consumption through dynamic voltage adjustment technology.

[0010] (a) Antenna design optimization: Unlike the traditional solution that uses a single dipole antenna, this solution uses a microstrip patch antenna array. Through array design, the operating frequency is set in the 5.8GHz ISM band to avoid the crowded 2.4GHz band, achieving a gain of 8dBi, while the traditional solution only achieves 3dBi. At the same time, the beamwidth is compressed to 30° to form a directional beam to achieve accurate wake-up of infrared cameras in specific areas.

[0011] (b) Low-power circuit reconfiguration: The traditional constant power supply is changed to dynamic bias technology. Through the voltage regulation circuit with duty cycle adjustment, the RFFE power supply voltage is reduced from 3.3V in the traditional solution to 0.9V in the standby stage, and the wake-up delay is strictly controlled within 50ms to meet the real-time requirements of field data acquisition scenarios, reduce standby power consumption and extend the overall battery life.

[0012] (2) Data storage unit:

[0013] In response to the characteristics of large amounts of field monitoring data and unstable transmission environments, this system achieves intelligent optimization and allocation of storage resources through multi-dimensional collaborative design.

[0014] Capacity planning based on sensor type: For infrared cameras at 1080p and 30fps, the data volume is 3.2 times higher than that of thermal imagers at 640x512 resolution and 25fps, using the following formula: Dynamically allocate storage space;

[0015] Cache mechanism for communication performance adaptation: Considering that the measured transmission efficiency of Wi-Fi 6 is 60% of the theoretical value of 1.2Gbps, the minimum cache capacity is set to... To ensure data integrity;

[0016] Power consumption and capacity balancing strategy: Based on the constraint that each TB SSD consumes approximately 5W of power, accounting for 15% of the total power consumption of the pod, through... This achieves a balanced optimization of battery life and storage.

[0017] Mechanical structure capacity-limiting design: It adopts a dual M.2 SSD slot layout within the standard pod size (200×150×100mm), with a physical capacity limit of twice the maximum single disk capacity, and controls the operating temperature of the solid-state drive through heat dissipation fins.

[0018] (3) ROS-based flight control algorithm:

[0019] The algorithm adopts a hierarchical distributed ROS architecture and achieves high-precision autonomous flight in complex field environments through environmental perception adaptive adjustment, multi-objective path optimization and dynamic energy management.

[0020] (a) Adaptive PID Controller: Employing a dynamic gain adjustment strategy and an anti-saturation integrator design, the controller achieves online real-time parameter adjustment via ROS nodes. This controller breaks through the fixed parameter mode of PID controllers, achieving high-precision and stable control in complex environments. Specifically, the dynamic gain adjustment adaptively adjusts the control parameters based on flight altitude; the control gain increases by 1% for every 100 meters increase in altitude, addressing the problem of enhanced wind disturbance at high altitudes. The anti-saturation integrator sets an error threshold; when the tracking error falls below the threshold, it performs integral accumulation to prevent actuator saturation and overshoot.

[0021] Environmental adaptive adjustment: Parameters are dynamically adjusted according to the real-time environment - when there is strong wind (wind speed >10m / s), the proportional gain increases by 20% and the derivative gain increases by 30%, improving the anti-disturbance capability; when near the ground (height <5m), the integral term is cleared and the output amplitude is limited; when the pod is fully loaded (>2kg), all gains are reduced to 80% of their own.

[0022] (b) Hybrid path planning algorithm: Based on the genetic algorithm framework, it deeply integrates multi-objective optimization and dynamic constraint processing mechanism, breaking through the limitation of traditional path planning that only considers distance.

[0023] Multi-objective fitness function: This function comprehensively calculates path length, wind drag coefficient, predicted battery consumption, and communication quality score. The calculation formula is as follows: Through dynamic weight adjustment ( , , , Achieve multi-dimensional balance optimization in complex environments to shorten flight distance while ensuring mission completion;

[0024] Dynamic constraint handling mechanism: Unreachable nodes are removed in real time for power constraints, auxiliary waypoints are inserted at the edge of communication blind spots to improve coverage, and the mutation probability is adaptively increased by 50% near no-fly zones to ensure that the path always meets physical limitations and safety requirements;

[0025] Energy-sensing crossover operator: An innovative design selects the crossover point with a positive correlation to the power level. When the power level is below 30%, it automatically generates a shortest path priority scheme to achieve synergistic optimization of energy efficiency and task completion rate.

[0026] (c) Energy-optimal control strategy: Based on the nonlinear discharge characteristics of the battery, a dynamic speed regulation mechanism is designed to achieve synergistic optimization of endurance and mission execution efficiency.

[0027] The system acquires the battery's State of Charge (SOC) status in real time. When the SOC is below 30%, it automatically triggers a speed reduction mode, adjusting the speed to 70% of the original trajectory to prioritize a safe return. When the battery is at high charge, it calculates the optimal speed that matches the current wind resistance and power system efficiency (usually corresponding to the peak range of the motor efficiency curve) to reduce energy consumption per unit distance.

[0028] (4) Two-phase wake-up mechanism:

[0029] To address the issues of sparse distribution of infrared cameras in the field and high power consumption of traditional continuous scanning, a two-stage wake-up strategy is proposed, which combines BLE signal pre-scanning with microwave directional fine wake-up. Furthermore, the transmit power is adaptively adjusted based on the communication distance to achieve a balance between energy consumption and wake-up reliability. First, a BLE periodic broadcast mode is used to broadcast the device identification code and basic status information. Then, a microstrip patch antenna array transmits a 16-bit device ID Gold code sequence for matching.

[0030] (5) System hardware composition:

[0031] This system adopts a design scheme that deeply integrates high-performance embedded architecture with dedicated hardware modules. The main control unit is equipped with the NVIDIA Jetson TX2 platform, which is responsible for running the ROS system and processing navigation decisions and communication tasks in real time. The communication module adopts a dual-mode combination of BLE and Wi-Fi 6, supporting 150-meter microwave directional wake-up and 1.2Gbps high-speed transmission. The storage unit is equipped with a 1TB SSD. The power supply uses two 6000mAh lithium batteries, combined with a dynamic voltage regulation circuit, to support 4 hours of continuous operation.

[0032] The advantages of this invention compared to the prior art are:

[0033] 1. This invention innovatively integrates a two-level communication mechanism of Bluetooth Low Energy pre-scanning and microwave directional wake-up, achieving long-distance, high-precision, and low-power device wake-up functionality. Compared to the traditional BLE single-mode solution with an effective range of 30 meters, a false wake-up rate of 15%, and a continuous power consumption of 3mW, this invention extends the effective wake-up distance to 150 meters. Furthermore, it utilizes correlation detection of Gold code sequences to reduce the false wake-up rate, and employs pulsed signal transmission to optimize average power consumption to 0.2mW. This significantly improves the reliability, coverage, and battery life of device connections in complex outdoor environments.

[0034] 2. This invention incorporates a dynamic gain adjustment strategy and an anti-saturation integrator design into the PID controller, enabling adaptive control under multiple abnormal operating conditions such as sudden wind interference, communication interruption, and sudden battery depletion. Compared to the limitations of traditional PID controllers, which suffer from a trajectory deviation of up to 2.1m under 8m / s wind, loss of control and crash after 30 seconds of communication interruption, and mission termination due to sudden battery depletion, this invention reduces the deviation under sudden wind interference to 0.3m through real-time parameter adjustment based on environmental perception. It automatically executes a contingency plan to recover after communication interruption and can still maintain operation at reduced frequency during sudden battery depletion, significantly enhancing the robustness and safety of the system.

[0035] 3. This invention improves upon the standard genetic algorithm by designing a hybrid path planning algorithm, which shortens the actual flight distance and increases the coverage of a single operation. Traditional genetic algorithms only consider distance, have a fixed population generation, cannot dynamically update paths, and do not consider battery power constraints. This invention integrates multiple factors such as wind field, terrain, and communication models to construct a multi-objective fitness function, designs dynamic constraint processing, and introduces energy sensing, supporting ROS online replanning and dynamic termination conditions. An energy-optimal control strategy is designed; compared to uniform power consumption, it utilizes the battery discharge curve to optimize flight speed, thereby extending endurance and improving mission completion rate, making it suitable for wildlife monitoring scenarios with multiple obstacles and strong interference.

[0036] 4. This invention employs a microstrip patch antenna array for the UAV pod and reconstructs a low-power circuit, filling the gap in existing UAV pods that lack a dedicated functional module for field data acquisition. A two-stage wake-up mechanism and a dual-mode communication unit solve the problem of inefficient data interaction with infrared cameras. Attached Figure Description

[0037] Figure 1 This is the overall flowchart of the present invention. Detailed Implementation

[0038] To enable those skilled in the art to better understand the embodiments of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and implementation methods.

[0039] like Figure 1 As shown, the specific implementation steps of the present invention are as follows:

[0040] 1. During the mission preparation phase, ground control station operators import the precise coordinate data of all infrared cameras within the target monitoring area, as well as geographical constraints: no-fly zones and communication blind spots.

[0041] 2. After the above data is uploaded to the UAV's onboard main control unit, the ROS system in the UAV's pod starts the path planning node, runs the hybrid path planning algorithm, and plans the basic data acquisition path based on the distribution of infrared cameras.

[0042] 3. During the flight of the UAV, the hybrid path planning algorithm will dynamically evaluate the comprehensive efficiency of each data acquisition path based on wind resistance, battery consumption and communication quality, and adjust the path with the best energy efficiency, the least wind resistance and the most stable communication coverage.

[0043] 4. During the drone's flight, unlike path optimization, the drone will make necessary trade-offs to path nodes based on constraints, prioritizing the normal completion of the mission: removing unreachable remaining nodes under power constraints; inserting auxiliary waypoints at the edge of communication blind spots when encountering them; and increasing the mutation rate by 50% when approaching no-fly zones, thus allowing the algorithm to avoid no-fly zones faster and more intelligently.

[0044] 5. The UAV collects data from all the infrared cameras it passes along the planned flight path. The specific steps for each data collection session include:

[0045] The drone arrives at the approximate airspace of the target area via the navigation system, then accurately positions itself within five meters of the target area and hovers there to initiate the data collection phase.

[0046] The drone pod first transmits a specific BLE signal using a microwave communication module, followed by a 16-bit Gold code sequence for the device ID. An infrared camera receives these signals for detection.

[0047] After successful authentication, a high-speed, stable Wi-Fi 6 connection is established between the drone and the infrared camera. The transmission protocol stack starts up, and high-definition video data stored in the infrared camera begins to be transmitted in batches at a theoretical rate of 1.2Gbps.

[0048] If a signal interruption occurs for more than 30 seconds during communication, the communication node will automatically trigger a reconnection mechanism. If three reconnections fail, the drone will automatically return to base. During communication or flight, if the battery level drops to the 30% warning threshold, the drone will adjust its flight speed to reduce. If the battery level further drops to 20%, the mission will be immediately and unconditionally terminated and the drone will return to base.

[0049] 6. After completing data collection at all preset monitoring points, or due to insufficient battery power, the drone will automatically return to its take-off and landing point. After the drone is recovered, staff can efficiently and securely download the massive amount of monitoring data stored on the pod's SSD to the ground workstation for subsequent browsing, analysis, and archiving via a gigabit Ethernet interface or a high-speed physical interface such as USB 3.0.

[0050] Through the above steps, this invention provides a drone pod data acquisition system based on a robot operating system. By deeply integrating directional microwave wake-up and ROS intelligent control system, it achieves stable and high-precision data acquisition from outdoor infrared camera networks, extends the effective wake-up distance to 150 meters, and provides up to 4 hours of continuous operation per session, significantly improving the task completion rate and providing efficient and reliable technical support for outdoor data acquisition.

Claims

1. A drone pod data acquisition system based on a robot operating system, characterized by, The application relates to a special-purpose pod for unmanned aerial vehicles (UAVs), which is equipped with a microwave wake-up module with an effective distance of 150 meters, a special-purpose data storage unit, an intelligent control system based on ROS, a two-stage wake-up mechanism, an anti-interference strategy, and an antenna design optimization. The pod is equipped with a special-purpose hardware, which integrates a microwave wake-up module supporting an effective distance of 150 meters. The module optimizes the antenna design and reconstructs a low-power circuit. In addition, the pod is equipped with a corresponding special-purpose data storage unit according to its own configuration. The special-purpose pod integrates an intelligent control system, which develops flight control algorithms based on ROS. Specifically, the system contains an adaptive proportional-integral-derivative (PID) controller and a hybrid path planning algorithm, which dynamically plans flight routes based on multi-source factors and collects data. The UAV pod uses a two-stage wake-up mechanism to establish a connection at the communication protocol layer, uses a dual-mode communication unit for data transmission, and designs an anti-interference strategy. The antenna design optimization uses a microstrip patch antenna array with a working frequency of 5.8 GHz ISM frequency band, and the beam width is compressed to 30° to realize directional wake-up.

2. The drone pod data collection system based on a robot operating system of claim 1, wherein: The low-power circuit reconstruction uses a dynamic bias technology based on a duty cycle voltage regulation circuit. When on standby, the radio frequency front end (RFFE) power supply is reduced to 0.9V, and the wake-up delay is controlled within 50ms.

3. The drone pod data collection system based on a robot operating system of claim 1, wherein: The storage unit capacity needs to be determined according to the pod configuration. First, consider the type of data collected by sensors, which needs to meet the formula: (Data type) * (Data size) <= (Storage unit capacity). ； Second, consider the communication module performance, i.e., the cache. Under the Wi-Fi 6 protocol, the theoretical rate is 1.2Gbps, and the actual transmission rate is 60%, i.e., 0.72Gbps, which needs to meet the formula: (Data type) * (Data size) <= (Storage unit capacity). ； Third, consider the power system capacity. Each 1TB storage unit consumes 15% of the total power consumption. The relationship between endurance and storage needs to be balanced, and the formula needs to be met: (Power system capacity) * (Endurance) >= (Storage unit capacity). ； Finally, consider the mechanical structure limitations of the UAV pod. The standard pod can accommodate up to 2 M.2 SSDs.

4. The drone pod data collection system based on a robot operating system of claim 1, wherein: The flight control algorithm developed based on ROS can realize dynamic gain adjustment strategy through adaptive PID controller; the hybrid path planning algorithm is improved based on standard genetic algorithm; and the energy optimal control strategy is formulated to optimize the cruise speed using the battery discharge curve.

5. The drone pod data collection system based on a robot operating system of claim 5, wherein: In the improved hybrid path planning algorithm, a multi-objective fitness function is designed to calculate the comprehensive fitness according to the path length, wind resistance, power, and communication quality. A dynamic constraint processing mechanism is added to select the corresponding processing strategy for different constraints, such as power, communication blind area, and no-fly zone. The energy-aware crossover operator is also included in the path planning algorithm, which makes the selection of crossover points positively related to the current power.

6. The drone pod data collection system based on a robotic operating system of claim 5, wherein: Communication delay and GPS front-end positioning drift are simulated in ROS.

7. The drone pod data collection system based on a robotic operating system of claim 1, wherein: The two-stage wake-up mechanism broadcasts BLE signals in the first stage and uses microwave directional beam scanning in the second stage to send a 16-bit device ID Gold code sequence. An adaptive power control algorithm is used in communication to adjust power consumption according to distance, maximizing power utilization.

8. The drone pod data collection system based on a robotic operating system of claim 1, wherein: The anti-interference strategy adjusts the UAV height for reconnection when the packet loss rate reaches 30% in communication, and returns automatically after three timeouts. A dynamic control parameter adjustment table is designed to use corresponding PID parameter adjustment strategies in different environments.

Images