Vehicle accident monitoring system and method in high-risk environment

Abstract

The invention relates to the technical field of vehicle accident monitoring systems, and discloses a vehicle accident monitoring system and method in a high-risk environment and a ground vehicle, the ground vehicle is equipped with a sensor suite, a local wireless transmitter, a parking apron and a plurality of unmanned aerial vehicles, the unmanned aerial vehicles are accommodated in the parking apron, and the unmanned aerial vehicles are connected with the sensor suite. And the control processor is configured to execute a shadow formation protocol, and at least one unmanned aerial vehicle is deployed to keep a continuous air trailing position behind the ground vehicle during the operation of the system. Aiming at the core pain point of the existing driving vehicle data recording scheme, the comprehensive breakthrough is realized: firstly, through data recording equipment and vehicle physical decoupling and dual storage redundancy, the data loss risk of a traditional airborne black box caused by vehicle damage is thoroughly avoided, and the accident data integrity is guaranteed;

Description

Technical Field

[0001] This invention relates to the field of vehicle accident monitoring system technology, specifically to a vehicle accident monitoring system and method for high-risk environments. Background Technology

[0002] With the rapid development of autonomous driving technology, autonomous vehicles (AVs) have been gradually applied in complex and high-risk scenarios such as tunnels and industrial mines. In such environments, vehicle operation faces potential safety hazards such as collisions, structural collapses, and battery fires. The tracing of causes and determination of responsibility after an accident rely on complete operational data. Currently, autonomous vehicles generally rely on onboard "black boxes" as core data recording devices, which store internal variable data such as speed and braking status collected by sensors to provide a basis for accident reconstruction.

[0003] However, existing data recording schemes have significant limitations in practical applications and are unable to meet the reliable data preservation requirements in high-risk environments: First, it is susceptible to physical damage: In catastrophic failure scenarios such as high-speed collisions, battery fires, or tunnel collapses, the autonomous vehicle itself is often severely damaged, and the onboard storage device may be crushed or burned, resulting in the complete loss of critical data before it is uploaded to the cloud, which cannot provide effective support for accident analysis. Secondly, there is a risk of mechanical structure failure: some proposed "ejection-type" black box solutions rely on mechanical actuators such as springs and explosive bolts to complete the ejection action under the complex G-force generated by the collision. However, the vehicle chassis is very prone to deformation during the collision, which may cause the ejection mechanism to jam, making it impossible to achieve effective detachment of the black box and data preservation; Third, it lacks key background information: the onboard black box can only record the internal operating variables of the vehicle and cannot capture the external environment and the entire collision process at the time of the accident. In other words, it lacks "third-person" external perspective data, which makes it difficult to fully restore the truth of the event during the accident reconstruction process and may affect the accuracy of liability determination.

[0004] In summary, existing data recording systems are either easily damaged due to their physical binding to the vehicle, or have potential failure risks due to their reliance on mechanical structures, and they cannot provide data from an external perspective, making it difficult to meet the safety data recording needs of autonomous vehicles in high-risk environments. Therefore, there is an urgent need for a solution with "proactive" data storage capabilities that can continuously synchronize vehicle operation data while maintaining physical isolation from hazard sources to ensure the integrity and security of accident data, providing comprehensive and reliable support for accident tracing and analysis. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a vehicle accident monitoring system and method for high-risk environments. It solves the problems that existing data recording systems are either easily damaged due to physical binding to the vehicle, or have potential failure risks due to reliance on mechanical structures, and cannot provide external perspective data, making it difficult to meet the safety data recording needs of autonomous vehicles in high-risk environments.

[0006] This invention provides the following technical solution: a vehicle accident monitoring system in a high-risk environment, comprising: a ground vehicle equipped with a sensor kit, a local wireless transmitter, and a helipad; Multiple unmanned aerial vehicles (UAVs) are housed within the helipad; A control processor configured to execute a "shadow formation protocol" that, during system operation, deploys at least one unmanned aerial vehicle to maintain a continuous aerial tail position behind the ground vehicle. The ground vehicle is configured to continuously transmit a mirror stream of critical telemetry data to the trailing unmanned aerial vehicle, which is configured to store the critical telemetry data in a rolling circular buffer.

[0007] Preferred technical solution one: The system is further configured to execute a "hot-swap rotation protocol", specifically including: Monitor the battery level of the trailing unmanned aerial vehicle; Launch a second fully charged drone from the tarmac before the battery of the trailing drone runs out; Synchronize the critical telemetry data stream to the second unmanned aerial vehicle; The original trailing unmanned aerial vehicle was landed on the tarmac for charging to maintain continuous external data redundancy.

[0008] Preferred technical solution two: The tail-following unmanned aerial vehicle is configured to detect "critical failure events" relative to ground vehicles, including collisions or signal loss. When such events are detected, the tail-following unmanned aerial vehicle performs the following operations: Permanently lock the data in the rolling circular buffer to non-volatile memory; Enter "Visit Hover Mode" to maintain visual contact with ground vehicles while keeping a safe physical distance; Send an emergency distress beacon.

[0009] Preferred technical solution three: The tailing unmanned aerial vehicle uses the "aerodynamic wake surfing" algorithm to position itself in the laminar flow region of the tunnel relative to the wake turbulence of ground vehicles.

[0010] Preferred technical solution four: A method for storing vehicle accident monitoring data under high-risk environments, comprising the following steps: Deploy drones so that they fly in formation behind autonomous vehicles; Establish high-bandwidth local wireless links between autonomous vehicles and drones; Continuously stream sensor data from autonomous vehicles to a temporary buffer on the drone; Detecting collision events involving autonomous vehicles; Save the contents of the temporary buffer on the drone to permanent storage to ensure that the data survives independently of the physical state of the autonomous vehicle.

[0011] Preferred technical solution four: The sensor kit includes an inertial measurement unit, millimeter-wave radar, high-definition camera and environmental sensor. The environmental sensor is used to collect temperature, humidity, dust concentration and visibility data in high-risk environments and simultaneously incorporate them into the mirror stream of key telemetry data.

[0012] Preferred technical solution five: The local wireless transmitter adopts a 5G millimeter wave communication module, supports anti-interference encrypted transmission, has a transmission delay of ≤100ms, and is equipped with a backup Bluetooth communication link, which automatically switches when the main link signal is interrupted, ensuring the continuity of telemetry data transmission.

[0013] Preferred technical solution six: The emergency distress beacon contains real-time location information of ground vehicles, key fault event types, and a thumbnail preview of data locked in non-volatile storage, which can be simultaneously sent to a preset monitoring platform, emergency rescue terminal, and management terminal associated with ground vehicles.

[0014] Preferred technical solution seven: The temporary buffer has a storage capacity of ≥100GB, supports data cyclic overwriting, and the overwriting priority follows the principle of "old data first overwriting". The buffer is also equipped with a data verification module to verify the integrity of the streaming data in real time and avoid data loss or corruption.

[0015] Preferred technical solution eight: The control processor is also configured to execute a "formation fault tolerance protocol". When the following unmanned aerial vehicle experiences a flight failure and cannot continue to follow, and there is no backup aircraft available for rotation, the faulty aircraft is automatically controlled to make an emergency landing at the nearest location, and the local data backup mechanism of the ground vehicle is simultaneously activated to save the key telemetry data to the ground vehicle's built-in non-volatile storage device.

[0016] Compared with existing technologies, this invention provides a vehicle accident monitoring system and method for high-risk environments, with the following beneficial effects: This invention achieves a comprehensive breakthrough in addressing the core pain points of existing vehicle data recording solutions through an innovative "shadow drone" system architecture and core mechanism design: First, by physically decoupling the data recording device from the vehicle and implementing dual storage redundancy, it completely avoids the risk of data loss due to vehicle damage associated with traditional airborne black boxes, ensuring the integrity of accident data; Second, by abandoning the mechanical ejection structure and adopting a UAV active tailing and seamless rotation mechanism, it eliminates the risk of mechanical failure, and combines aerodynamic optimization algorithms to reduce energy consumption, significantly improving system reliability and endurance; Third, by utilizing the UAV's aerial "witness" perspective, it supplements the external environment and process images of the accident, enriching the data dimensions and helping to accurately reconstruct the truth of the accident; Fourth, after the accident, it automatically locks the data, sends location beacons, and transmits real-time on-site information, balancing data tracing and emergency rescue, improving rescue efficiency.

[0017] Overall, this invention significantly enhances the data security, system stability, and comprehensiveness of accident analysis for vehicles driven in high-risk environments. It also has strong scenario adaptability, providing key support for the large-scale application of driving technology in complex scenarios. Attached Figure Description

[0018] Figure 1 This is a side view schematic diagram of the "shadow formation" of the present invention; Figure 2 This is the logic diagram of the "heat exchange rotation" of the present invention; Figure 3 This is a timing diagram illustrating the "collision response" of the present invention; Figure 4 This is a flowchart illustrating the normal operation and heat exchange cycle of the system according to the present invention; Figure 5 This is a flowchart illustrating the accident data storage process of the present invention. Detailed Implementation

[0019] Please see Figure 1-5 , Example 1: A vehicle accident monitoring system for high-risk environments, comprising: A ground vehicle equipped with a sensor suite, a local wireless transmitter, and a helipad; Multiple unmanned aerial vehicles (UAVs) are housed on the tarmac. The control processor is configured to execute the "shadow formation protocol," which deploys at least one unmanned aerial vehicle to maintain a continuous aerial tail position behind the ground vehicle during system operation. The ground vehicle is configured to continuously transmit a mirror stream of critical telemetry data to the trailing unmanned aerial vehicle, which is configured to store the critical telemetry data in a rolling circular buffer.

[0020] The system is further configured to execute a "hot-swap rotation protocol", which specifically includes: Monitor the battery level of the trailing unmanned aerial vehicle; Launch a second fully charged drone from the tarmac before the battery of the trailing drone runs out; Synchronize the critical telemetry data stream to the second unmanned aerial vehicle; The original trailing unmanned aerial vehicle was landed on the tarmac for charging to maintain continuous external data redundancy.

[0021] Example 2: The difference between this example and Example 1 is that the trailing unmanned aerial vehicle is configured to detect "critical failure events" relative to ground vehicles. "Critical failure events" include collisions or signal loss. When such events are detected, the trailing unmanned aerial vehicle performs the following operations: Permanently lock the data in the rolling circular buffer to non-volatile memory; Enter "Visit Hover Mode" to maintain visual contact with ground vehicles while keeping a safe physical distance; Send an emergency distress beacon.

[0022] Example 3: The difference between this example and Example 1 is that the trailing unmanned aerial vehicle uses the "aerodynamic wake surfing" algorithm to position itself in the laminar flow region of the tunnel relative to the wake turbulence of ground vehicles.

[0023] Example 4: The difference between this example and Example 1 is that the method for storing vehicle accident monitoring data in a high-risk environment includes the following steps: Deploy drones so that they fly in formation behind autonomous vehicles; Establish high-bandwidth local wireless links between autonomous vehicles and drones; Continuously stream sensor data from autonomous vehicles to a temporary buffer on the drone; Detecting collision events involving autonomous vehicles; Save the contents of the temporary buffer on the drone to permanent storage to ensure that the data survives independently of the physical state of the autonomous vehicle.

[0024] Example 5: The difference between this example and Example 1 is that the sensor kit includes an inertial measurement unit, a millimeter-wave radar, a high-definition camera, and an environmental sensor. The environmental sensor is used to collect temperature, humidity, dust concentration, and visibility data in high-risk environments and simultaneously incorporate them into the mirror stream of key telemetry data.

[0025] Example 6: The difference between this example and Example 1 is that the local wireless transmitter uses a 5G millimeter-wave communication module, supports anti-interference encrypted transmission, has a transmission delay of ≤100ms, and is equipped with a backup Bluetooth communication link, which automatically switches when the main link signal is interrupted, ensuring the continuity of telemetry data transmission.

[0026] Example 7: The difference between this example and Example 1 is that the emergency distress beacon contains real-time location information of the ground vehicle, key fault event types, and a thumbnail preview of data locked in non-volatile storage, which can be simultaneously sent to a preset monitoring platform, emergency rescue terminal, and management terminal associated with the ground vehicle.

[0027] Example 8: The difference between this example and Example 1 is that the temporary buffer has a storage capacity of ≥100GB, supports data cyclic overwriting, follows the principle of "old data first overwriting" for overwriting priority, and is equipped with a data verification module to verify the integrity of streaming data in real time to avoid data loss or corruption.

[0028] Example 9: The difference between this example and Example 1 is that the control processor is also configured to execute a "formation fault tolerance protocol". When the following unmanned aerial vehicle experiences a flight failure and cannot continue to follow and there is no backup aircraft available for rotation, the faulty aircraft is automatically controlled to make an emergency landing at the nearest location, and the local data backup mechanism of the ground vehicle is simultaneously activated to save the key telemetry data to the ground vehicle's built-in non-volatile storage device.

[0029] In this embodiment, existing data recording systems are either easily damaged due to their physical binding to the vehicle, or have potential failure risks due to reliance on mechanical structures, and cannot provide external perspective data, making it difficult to meet the safety data recording requirements of autonomous vehicles in high-risk environments. Therefore, there is an urgent need for a solution with "proactive" data storage capabilities that can continuously synchronize vehicle operation data while maintaining physical isolation from hazard sources to ensure the integrity and security of accident data, providing comprehensive and reliable support for accident tracing and analysis.

[0030] In summary, in specific implementation, I. System Overall Architecture Deployment The core components of this system include a main ground vehicle, multiple unmanned aerial vehicles (UAVs), a control processor, and supporting communication and storage modules. The deployment and configuration of each component are as follows: Main ground vehicle configuration: The ground vehicle is equipped with a complete sensor suite, a local high-bandwidth wireless transmitter, and a multi-compartment parking apron. The sensor suite includes LiDAR, high-definition cameras, inertial measurement units, etc., used to collect raw perception data and key telemetry data such as control input signals (e.g., throttle, braking, steering commands) during vehicle operation; the local wireless transmitter adopts WiGig or 60GHz millimeter-wave technology, supporting a transmission bandwidth of ≥10Gbps to ensure real-time, zero-latency data transmission; the multi-compartment parking apron has at least two independent compartments, each with a built-in wireless charging coil (300mm in diameter, 50W power), compartment dimensions of 600×600×250mm, spacing of 300mm, and an infrared guidance mechanism at the compartment entrance with a guidance accuracy of ±10mm, used for UAV docking, charging, and rapid dispatch.

[0031] UAV Configuration: The UAV is constructed from a composite material of carbon fiber and engineering plastics, with a wingspan of 900mm, a fuselage length of 600mm, and a height of 200mm. It is equipped with an optical camera (35° field of view, supporting 4K resolution, 60fps frame rate, and infrared night vision), non-volatile SSD storage (capacity ≥1TB, read / write speed ≥500MB / s), a high-bandwidth receiver, and an emergency beacon transmitter (transmission frequency 433MHz, transmission distance ≥5km). The UAV has a built-in RAM rolling circular buffer, defaulted to storing the most recent 5 minutes of telemetry data, with the storage duration adjustable according to actual needs.

[0032] Control processor deployment: The control processor is integrated inside the ground vehicle and communicates with the vehicle control system, sensor suite, wireless transmitter, and helipad control module. It also establishes a real-time control channel with the UAV via a wireless link. The control processor is pre-loaded with the "Shadow Formation Protocol," "Hot-Swap Rotation Protocol," and "Aerodynamic Optimization Algorithm," and is responsible for coordinating the UAV's flight status, data transmission synchronization, and emergency response actions.

[0033] II. Execution Flow of Core Protocols and Algorithms (a) Execution of Shadow Formation Protocol After the system starts up, the control processor first completes self-test and equipment initialization. After confirming that the sensor kit, wireless communication module and apron status are normal, it sends a takeoff command to one of the UAVs in the cabin.

[0034] Upon receiving the command, the UAV takes off from the tarmac and, guided by the control processor, flies to a pre-set tailing position 50 meters behind the ground vehicles. This position is aligned with the top of the vehicles, forming a stable "shadow formation." During this process, the UAV adjusts its attitude in real time using position and speed data transmitted from the ground vehicles, ensuring a hovering accuracy of ±0.5m (relative to the center of the vehicle's rear).

[0035] The sensor suite of the ground vehicle continuously collects key telemetry data and broadcasts it continuously to the trailing UAV in the form of a data mirror stream via a local high-bandwidth wireless link. After receiving the data, the UAV stores it in a RAM rolling circular buffer. The buffer adopts a circular storage mechanism of "new data overwrites old data" to always retain the most recent valid data.

[0036] (ii) Implementation of the heat exchange rotation agreement To address the battery life limitations of a single UAV (20-30 minutes per flight), the system employs a "tag-team" logic to achieve seamless UAV rotation, ensuring 100% time coverage for data backup. The specific process is as follows: The control processor monitors the battery level of the UAV currently in a tailing state (hereinafter referred to as "active UAV") in real time and sets the remaining battery level of 20% as the low battery threshold.

[0037] When the battery power of the active UAV drops to the low power threshold, the control processor immediately sends a takeoff command to the fully charged standby UAV (hereinafter referred to as "standby UAV") on the tarmac. The standby UAV takes off from its cabin and flies towards the trailing position of the active UAV along a preset trajectory (an arc trajectory with a radius of 3000mm and a flight speed of 2m / s).

[0038] During the flight of the backup UAV, a high-bandwidth data link is established with the ground vehicle. The buffer data and current flight parameters of the active UAV are synchronized through the "Make-Before-Break" logic, with a synchronization delay of ≤1ms, to ensure no data loss.

[0039] Once the backup UAV reaches the trailing position and completes data synchronization, the control processor sends a landing command to the active UAV. The active UAV returns to an available bay on the apron along an arc-shaped landing path (radius 3000mm, flight speed 2m / s). The bay's infrared guidance mechanism guides the UAV to a precise docking, and then wireless charging is initiated.

[0040] The backup UAV officially takes over the tailing task and becomes the new active UAV. The system enters the next round of power monitoring and rotation cycle. The entire rotation process takes ≤10 seconds, and the vehicle does not need to stop driving.

[0041] (III) Application of Aerodynamic Optimization Algorithms In enclosed or semi-enclosed high-risk environments such as tunnels and industrial mines, the movement of ground vehicles generates wake turbulence, increasing the UAV's flight drag and battery consumption. To address this, the UAV is equipped with an "aerodynamic wake surfing" algorithm, the specific execution process of which is as follows: UAVs collect real-time data such as the speed of ground vehicles and the dimensions of tunnel cross-sections (width and height) through their own sensors, and transmit the data to the control processor.

[0042] The control processor calculates the laminar flow zone within the tunnel (usually located at the top of the tunnel or the shoulder of the lane, known as the "Laminar Flow Pocket") using algorithms and sends flight vector commands to the UAV.

[0043] The UAV adjusts its flight attitude and position according to instructions, accurately positioning itself within the laminar flow region to minimize energy loss from combating wake turbulence and extend its single flight endurance.

[0044] III. Implementation of the Accident Response Mechanism When a critical failure event such as a collision between ground vehicles or structural collapse occurs, the system immediately activates the accident response mechanism, the specific procedure of which is as follows: Fault event detection: The UAV detects fault events in two ways: first, by receiving collision signals sent by ground vehicles (such as electrical signals triggered by airbag deployment); second, by monitoring the data transmission link status. If a signal interruption lasting ≥0.3s occurs, it is determined to be a fault event that may cause damage to the vehicle.

[0045] Data locking and preservation: After the UAV detects a fault event, the processor immediately executes the "write lock" instruction, which permanently writes the data stored in the rolling circular buffer before the collision (the last 5 minutes by default) to the non-volatile SSD storage within 0.1s, ensuring that the data is not overwritten or lost.

[0046] Witness mode activated: The UAV adjusts its flight position and flies to a safe hovering area 100 meters away from the accident vehicle, maintaining visual contact with the accident vehicle. At the same time, it activates the optical camera to record external perspective videos such as the vehicle's kinematic state after the collision and the onset of fire / smoke, providing "third-person" witness evidence for accident analysis.

[0047] Emergency distress beacon transmission: The UAV's emergency beacon transmitter is immediately activated, sending a distress signal containing information such as the accident location (obtained via GPS positioning) and vehicle identification to a radius of 5km. The signal continuously reports its location to emergency personnel and the back-end management system until rescue personnel arrive or the system receives a reset command.

[0048] IV. Data Storage and Synchronization Guarantee Data transmission synchronization: Data transmission between ground vehicles and UAVs adopts a synchronous mirroring protocol. While the ground vehicle stores data locally, it mirrors the data to the UAV in real time through a high-bandwidth link with a transmission latency of ≤10ms, ensuring that the data stored in the UAV is completely consistent with the local data in the vehicle.

[0049] Data security assurance: UAV's non-volatile SSD storage uses encrypted storage to prevent data from being illegally tampered with or stolen; after the data is saved in an incident, it supports uploading backups to the cloud management platform via a wireless link, further enhancing data security.

[0050] Buffer Management: The storage duration of the rolling circular buffer can be flexibly configured by controlling the processor to meet the data analysis needs of different scenarios, while avoiding performance loss due to excessive storage capacity.

Claims

1. A vehicle accident monitoring system for high-risk environments, characterized in that: include: A ground vehicle equipped with a sensor suite, a local wireless transmitter, and a helipad; Multiple unmanned aerial vehicles (UAVs) are housed within the helipad; A control processor configured to execute a "shadow formation protocol" that, during system operation, deploys at least one unmanned aerial vehicle to maintain a continuous aerial tail position behind the ground vehicle. The ground vehicle is configured to continuously transmit a mirror stream of critical telemetry data to the trailing unmanned aerial vehicle, which is configured to store the critical telemetry data in a rolling circular buffer.

2. The vehicle accident monitoring system in a high-risk environment according to claim 1, characterized in that: The system is further configured to execute a "hot-swap rotation protocol", specifically including: Monitor the battery level of the trailing unmanned aerial vehicle; Launch a second fully charged drone from the tarmac before the battery of the trailing drone runs out; Synchronize the critical telemetry data stream to the second unmanned aerial vehicle; The original trailing unmanned aerial vehicle was landed on the tarmac for charging to maintain continuous external data redundancy.

3. The vehicle accident monitoring system in a high-risk environment according to claim 1, characterized in that: The trailing unmanned aerial vehicle is configured to detect "critical failure events" relative to ground vehicles, including collisions or signal loss. When such events are detected, the trailing unmanned aerial vehicle performs the following operations: Permanently lock the data in the rolling circular buffer to non-volatile memory; Enter "Visit Hover Mode" to maintain visual contact with ground vehicles while keeping a safe physical distance; Send an emergency distress beacon.

4. The vehicle accident monitoring system in a high-risk environment according to claim 1, characterized in that: The trailing unmanned aerial vehicle uses an "aerodynamic wake surfing" algorithm to position itself within the laminar flow region of the tunnel relative to the wake turbulence of ground vehicles.

5. A method for storing vehicle accident monitoring data in high-risk environments, characterized in that, Includes the following steps: Deploy drones so that they fly in formation behind autonomous vehicles; Establish high-bandwidth local wireless links between autonomous vehicles and drones; Continuously stream sensor data from autonomous vehicles to a temporary buffer on the drone; Detecting collision events involving autonomous vehicles; Save the contents of the temporary buffer on the drone to permanent storage to ensure that the data survives independently of the physical state of the autonomous vehicle.

6. The vehicle accident monitoring system in a high-risk environment according to claim 1, characterized in that: The sensor suite includes an inertial measurement unit, millimeter-wave radar, high-definition camera, and environmental sensor. The environmental sensor is used to collect temperature, humidity, dust concentration, and visibility data in high-risk environments and simultaneously incorporate them into a mirror stream of key telemetry data.

7. The vehicle accident monitoring system in a high-risk environment according to claim 1, characterized in that: The local wireless transmitter uses a 5G millimeter-wave communication module, supports anti-interference encrypted transmission, has a transmission latency of ≤100ms, and is equipped with a backup Bluetooth communication link that automatically switches when the main link signal is interrupted, ensuring the continuity of telemetry data transmission.

8. The vehicle accident monitoring system in a high-risk environment according to claim 3, characterized in that: The emergency distress beacon contains real-time location information of ground vehicles, key fault event types, and a thumbnail preview of data locked in non-volatile storage, which can be simultaneously sent to a preset monitoring platform, emergency rescue terminal, and management terminal associated with the ground vehicles.

9. The method for storing vehicle accident monitoring data in high-risk environments according to claim 5, characterized in that: The temporary buffer has a storage capacity of ≥100GB, supports data cyclic overwriting, and the overwriting priority follows the principle of "old data first overwriting". The buffer is also equipped with a data verification module to verify the integrity of the streaming data in real time and avoid data loss or corruption.

10. A vehicle accident monitoring system in a high-risk environment according to claim 1, characterized in that: The control processor is also configured to execute a "formation fault tolerance protocol". When the trailing unmanned aerial vehicle experiences a flight malfunction and cannot continue to follow, and there is no backup aircraft available for replacement, the malfunctioning aircraft is automatically controlled to make an emergency landing at the nearest location. At the same time, the local data backup mechanism of the ground vehicle is activated to save the key telemetry data to the ground vehicle's built-in non-volatile storage device.

Images