Fire-fighting detection unmanned vehicle driven by multi-degree-of-freedom mechanical arm

Abstract

The invention discloses a firefighting detection unmanned vehicle driven by a multi-degree-of-freedom mechanical arm, and belongs to the technical field of firefighting rescue equipment. The unmanned vehicle comprises a high-strength explosion-proof vehicle body, an explosion-proof direct-drive walking mechanism, a multi-degree-of-freedom operation mechanical arm, a multi-source environment sensing system and an intelligent control system. The mechanical arm has ten degrees of freedom, a six-dimensional force sensor is arranged at a key joint, a quick-change clamp is arranged at the tail end, and fine operations such as door opening and forcible entry can be achieved. The sensing system integrates a laser radar, a binocular camera and a gas sensor and is used for three-dimensional mapping and reconnaissance. The control system can automatically avoid obstacles and drive the mechanical arm to carry out smooth and accurate micro-motion adjustment by fusing vision and force information. The system can replace personnel to enter an extreme dangerous fire scene to execute reconnaissance and disposal tasks, has high maneuverability, environmental perceptivity, fine operation safety and multi-task adaptability, and remarkably improves rescue efficiency and safety.

Description

Technical Field

[0001] This invention relates to the field of fire rescue equipment technology, specifically to a fire detection unmanned vehicle driven by a multi-degree-of-freedom robotic arm. Background Technology

[0002] In fires occurring in complex environments such as petrochemical plants, underground utility tunnels, and large warehouses, the environment is typically characterized by high temperatures, dense smoke, a risk of collapse, and the potential presence of toxic and harmful gases. Directly dispatching firefighters deep into these environments for reconnaissance and response poses an extreme risk to their lives.

[0003] Currently, most firefighting robots on the market are remotely controlled tracked mobile platforms equipped with high-flow-rate fire monitors, primarily focused on powerful water jetting for fire suppression at external locations or entrances. These robots have significant limitations: First, they are functionally limited and cannot perform critical tactical maneuvers such as opening doors, breaking through obstacles, or close-range sampling; second, they lack precise operational capabilities, with robotic arms (if equipped) typically having limited degrees of freedom, clumsy movements, and an inability to flexibly adjust their posture in confined spaces; finally, they lack force sensing and compliant control capabilities, making them highly susceptible to causing secondary damage to precision equipment or building structures during contact operations, and even leading to jamming or collisions that cause them to malfunction.

[0004] For other specific rescue scenarios, such as forest fires, existing technologies also suffer from insufficient functional adaptability. Chinese utility model patent CN221874185U discloses a modular intelligent engineering vehicle that integrates a vehicle body, a micro-robot main body, and multiple functional modules for firefighting, logging, tree planting, and reconnaissance. This patented comprehensive rescue platform, designed for open terrain environments, focuses its functional modules and core tasks on field operations. Its mechanical structure is relatively large or specialized, making it unsuitable for industrial fire scenarios.

[0005] Therefore, there is still a need for specialized equipment that combines extremely high environmental tolerance, high mobility and flexibility, precise and dexterous operation capabilities, and powerful multimodal perception capabilities to replace personnel in performing critical tasks such as reconnaissance, sampling, and demolition in extremely dangerous areas. Thus, there is an urgent practical need to develop an unmanned firefighting reconnaissance and rescue system that can simultaneously meet the above stringent requirements. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent fire-fighting unmanned vehicle that integrates high mobility, environmental perception, precise operation and multi-task collaboration capabilities, so as to replace or assist firefighters in performing the most dangerous tasks and improve rescue efficiency and safety.

[0007] The technical solution of this invention: a fire detection unmanned vehicle driven by a multi-degree-of-freedom robotic arm, comprising... The chassis module includes a sealed explosion-proof body, hollow sealed anti-static rubber tires located at the bottom of the body, and a built-in lithium battery pack. The multi-degree-of-freedom operation module is located at the upper front of the vehicle body. It includes an aluminum alloy double-section robotic arm with 10 degrees of freedom. The surface of the robotic arm is coated with a high-temperature resistant and corrosion-resistant coating. Its end is connected to a quick-change gripper, and a six-dimensional force sensor is installed at its key joints. The environmental perception and communication module integrates a multi-functional gas sensor, lidar, high-definition explosion-proof PTZ camera and explosion-proof antenna at the rear of the vehicle, and a binocular depth camera at the front of the vehicle, which together are used to realize environmental modeling, reconnaissance and data transmission.

[0008] Furthermore, the robotic arm includes a base fixed to the vehicle body, a first robotic arm section, and a second robotic arm section; The first robotic arm is pivotally connected to the base via a rotary table and is driven by a rotary servo motor to achieve horizontal rotation. The first robotic arm is driven by the first servo motor. The second robotic arm is connected to the end of the first robotic arm and is driven by a second servo motor. The end of the second robotic arm is connected to a clamp, and a fixed camera is installed near the clamp.

[0009] Furthermore, the fixture includes a mounting base, a pair of mirror-symmetrically arranged bent grippers, and a pair of connectors. The mounting base is provided with mounting ears and a fixed shaft. One end of the two bent grippers is hinged to the mounting ears by a first locking pin and a third locking pin (19), respectively. The bent parts of the two bent grippers are connected to the connectors. The other ends of the pair of connectors are hinged to the end of the fixed shaft by a second locking pin (17), thereby forming a parallel four-bar linkage opening and closing mechanism to realize the synchronous opening and closing movement of the two grippers.

[0010] Furthermore, the gripper is fixedly connected to the end of the second robotic arm via its mounting base, and the gripper adopts a modular design, which can be quickly replaced according to different task requirements such as demolition and grasping.

[0011] Furthermore, the vehicle body is a high-strength carbon steel sealed explosion-proof structure that can withstand temperatures up to 800℃, and the wheels are driven by explosion-proof motors; the 200Ah / 48V lithium battery pack inside the vehicle supports ≥4 hours of operation and ≤1.5 hours of fast charging.

[0012] Furthermore, each joint of the robotic arm is equipped with a servo motor and an angle encoder coaxially connected to it. The servo motor and the angle encoder constitute a position closed-loop control system, which can drive the end gripper of the robotic arm to achieve 360-degree omnidirectional positioning and continuous attitude adjustment in three-dimensional space.

[0013] Furthermore, a first six-dimensional force sensor associated with the first servo motor is provided at the pivot connection between the first robotic arm and the base, and a second six-dimensional force sensor associated with the second servo motor is provided at the drive joint between the second robotic arm and the first robotic arm.

[0014] Furthermore, the six-dimensional force sensor acquires X / Y / Z axis force and torque signals in real time, with a measurement accuracy of ≤±0.5%FS and a preset force feedback threshold within the range of 10-100N.

[0015] An intelligent control system for controlling a fire detection unmanned vehicle driven by a multi-degree-of-freedom robotic arm is disclosed. The control system adopts a hierarchical architecture of an on-board industrial control computer and an embedded real-time controller. The system executes as follows: it fuses data from a six-dimensional force sensor with visual data from a camera to generate information on the distance to environmental obstacles and the contact force of the robotic arm. When the obstacle distance is ≤100cm or the contact force is ≥ a preset threshold, it autonomously generates obstacle avoidance control commands and drives the robotic arm to adjust its posture to achieve obstacle avoidance.

[0016] Furthermore, the data fusion processing is based on a weighted Kalman filter algorithm, which dynamically allocates weights to visual data and force sensing data, and outputs key information such as the distance to environmental obstacles and the magnitude of contact force after fusion. The attitude adjustment of the driving robotic arm adopts a fuzzy adaptive PID control algorithm; The algorithm takes the posture error (e) and error change rate (ec) of the robotic arm as input, adjusts the PID control parameters online through a 5×5 fuzzy rule matrix, and updates the control command with a period of 20ms. Ultimately, the robotic arm is controlled to adjust its posture at a micro-motion speed of ≤5mm / s, and the positioning error of its end effector is controlled within ±2mm. The beneficial effects of this invention are: 1. This invention improves the safety of rescue operations by enabling machines to replace human labor. The unmanned vehicle adopts a high-strength carbon steel sealed explosion-proof body that can withstand temperatures up to 800°C and is equipped with an explosion-proof direct drive system. This allows it to replace firefighters and directly enter the core area of ​​extreme fire scenes with high temperatures, dense smoke, toxic substances, and risks of collapse to perform the most dangerous reconnaissance and initial response tasks, fundamentally avoiding the life-threatening dangers faced by firefighters.

[0017] 2. This invention enables precise operation and safe interaction in complex environments. Through a double-segment robotic arm with 10 degrees of freedom and high-precision six-dimensional force sensors integrated into key joints, it achieves micro-operations with millimeter-level precision, completely avoiding secondary damage to equipment or self-jamming failure caused by brute force operation of traditional robotic arms.

[0018] 3. This invention constructs a comprehensive, multi-dimensional environmental perception and information advantage, integrating lidar, binocular depth cameras, explosion-proof PTZ cameras, multi-functional gas sensors, and explosion-proof communication antennas to form a multi-source perception system. This enables remote commanders to obtain real-time 3D maps of the fire scene, obstacle locations, toxic gas concentrations, and high-definition operational footage, providing comprehensive and real-time information support for scientific decision-making that is unparalleled by traditional manual reconnaissance.

[0019] 4. This invention achieves a transition from remote control to autonomous assistance through intelligent control. Based on a hierarchical architecture and advanced algorithms such as fuzzy adaptive PID, the control system can integrate visual and force information to autonomously make obstacle avoidance decisions and perform compliant posture adjustments. This significantly reduces operational complexity, allowing operators to focus on high-level task instructions rather than tedious low-level operations, thereby improving overall work efficiency and task success rate. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the specific embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the overall structure of a fire detection unmanned vehicle driven by a multi-degree-of-freedom robotic arm; Figure 2 This is a schematic diagram of the recovery of a fire detection and reconnaissance unmanned vehicle driven by a robotic arm with multiple degrees of freedom; Figure 3 This is a schematic diagram of the robotic arm structure; Figure 4 This is a schematic diagram of the fixture structure; Figure 5 This is a schematic diagram of the horizontal movement range of the robotic arm; Figure 6 This is a schematic diagram of the vertical movement range of the robotic arm; Figure 7 This is a block diagram of the PID adaptive control algorithm.

[0022] Reference numerals: 1-Vehicle body; 2-All-in-one gas sensor; 3-Tire; 4-Binocular depth camera; 5-Robotic arm; 6-LiDAR; 7-High-definition explosion-proof PTZ camera; 8-Explosion-proof antenna; 9-First six-dimensional force sensor; 10-Rotary servo motor; 11-First servo motor; 12-Second servo motor; 13-Second six-dimensional force sensor; 14-Fixed camera; 15-Gripper; 16-First locking pin; 17-Second locking pin; 18-Third locking pin; 19-Gripper; 20-First section of robotic arm; 21-Second section of robotic arm. Detailed Implementation

[0023] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. However, it should not be construed that the scope of the subject matter of the present invention is limited to the following embodiments. All modifications, substitutions and alterations made based on ordinary technical knowledge and common practices in the art without departing from the above-described technical concept of the present invention are included within the scope of the present invention.

[0024] The present invention relates to a multi-degree-of-freedom robotic arm-driven unmanned fire detection vehicle, with reference to... Figure 1 and Figure 2 Mainly includes: Chassis: The vehicle body adopts a sealed explosion-proof design, with a high-strength carbon steel outer shell that can withstand 800℃ for 30 minutes, achieving a protection level ≥IP67. The wheels use hollow, sealed, anti-static rubber tires, driven directly by an explosion-proof motor, simplifying the transmission mechanism, improving transmission efficiency, and providing strong passability and stability. It has a built-in 200Ah / 48V lithium battery pack, providing ≥4 hours of continuous operation. The power module has overcharge, over-discharge, and short-circuit protection functions, and fast charging time ≤1.5 hours.

[0025] Multi-DOF robotic arm system: Utilizing an aluminum alloy double-section arm with multiple joints, coated with a high-temperature resistant and corrosion-resistant coating, it possesses 10 degrees of freedom. The folding joints, controlled by servo motors and angle encoders, form a closed-loop control system, enabling the end effector to achieve 360-degree omnidirectional positioning and flexible attitude adjustment in complex spaces.

[0026] Six-dimensional force sensor: Installed on each joint of the robotic arm, it collects X / Y / Z axis force and torque signals in real time, with a measurement accuracy of ≤±0.5%FS, and the force feedback threshold can be preset in the range of 10-100N.

[0027] The vision and environmental perception system includes a lidar (for SLAM mapping and navigation), a binocular depth camera (for 3D vision servoing and target recognition), a high-definition explosion-proof PTZ camera (for monitoring the environment around the vehicle), a fixed camera (for monitoring the end effector of the robotic arm), and an all-in-one gas sensor.

[0028] End effector: Employs a mechanically locking quick-change mechanism to support rapid switching of demolition tools (such as demolition shears).

[0029] Control system: It adopts a layered architecture, including an on-board industrial computer (responsible for high-level decision-making, path planning and vision processing) and an embedded real-time controller (responsible for low-level servo control, sensor data fusion and compliant control algorithm execution).

[0030] 1) Data preprocessing: The six-dimensional force sensor data is filtered by moving average to remove noise, and the visual images are subjected to distortion correction and feature extraction.

[0031] 2) Multi-source information fusion: Based on the weighted Kalman filter algorithm, the weights of visual data and force sensing data are dynamically allocated, and the fused data outputs key information such as the distance to environmental obstacles and the magnitude of contact force.

[0032] 3) Obstacle avoidance decision: When the detected distance is ≤100cm or the contact force is ≥preset threshold, the algorithm generates an obstacle avoidance path and prioritizes the movement direction without obstacles.

[0033] 4) Attitude Adjustment: Output control commands to the servo motor, driving the robotic arm to adjust its attitude at a fine-tuning speed of ≤5mm / s to achieve compliant obstacle avoidance. Using the robotic arm attitude error e and error change rate ec as inputs, output PID parameter correction values ​​ΔKp, ΔKi, and ΔKd to establish a 5×5 fuzzy rule matrix. Update the PID parameters every 20ms to ensure the robotic arm attitude control error is ≤±2mm.

[0034] Example 1: A fire detection unmanned vehicle driven by a multi-degree-of-freedom robotic arm. The unmanned vehicle adopts a modular design and integrates four major functional modules: protection, walking, operation and perception.

[0035] High-strength protection and mobility module: Serving as the mobile base, the core is a sealed, explosion-proof vehicle body 1 made of high-strength carbon steel. This body can withstand temperatures up to 800℃ for 30 minutes, achieving an IP67 protection rating, ensuring survivability in extreme fire conditions. The underside of the body is equipped with four hollow, sealed, anti-static rubber tires 3, directly driven by explosion-proof motors, providing excellent passability and stability. The vehicle body has a built-in 200Ah / 48V lithium battery pack, supporting continuous operation for ≥4 hours and capable of rapid charging within 1.5 hours, ensuring efficient deployment.

[0036] Highly flexible operating module: The multi-degree-of-freedom robotic arm 5, located at the front of the vehicle body 1, is the core operating unit of this vehicle. For example... Figure 3 As shown, the robotic arm adopts an aluminum alloy double-segment design, has 10 degrees of freedom of motion, and is coated with a high-temperature resistant and corrosion-resistant coating. (Refer to...) Figure 5 and Figure 6 The first robotic arm 20 is driven by a rotary servo motor 10 to achieve horizontal translation within 70°, and by a first servo motor 11 to drive pitch; the second robotic arm 21 is driven by a second servo motor 12. The robotic arm can achieve a vertical rotation of 142°. To facilitate precise operation, a first six-dimensional force sensor 9 and a second six-dimensional force sensor 13 are respectively installed at the drive joints of the first and second robotic arms. The end of the robotic arm 5 is connected to a gripper 15, which in this embodiment is a parallel four-bar linkage gripper, such as... Figure 4As shown, the device consists of a pair of mirror-symmetrical bent grippers 19 that are hinged to the base via a first locking pin 16, a second locking pin 17, and a third locking pin 18, enabling synchronous opening and closing. The grippers 19 can also be quickly replaced according to task requirements.

[0037] All-around environmental perception module: This module provides observation and detection data for the unmanned vehicle. A binocular depth camera 4 is installed at the front of the vehicle body 1 for 3D vision recognition and servo control; the rear integrates a LiDAR 6 for simultaneous localization and mapping (SLAM), a high-definition explosion-proof PTZ camera 7 for panoramic environmental monitoring, and a multi-functional gas sensor 2 for detecting toxic and harmful gases. In addition, a fixed camera 14 is located near the end effector 15 of the robotic arm for real-time monitoring of the operation process. All data is transmitted via an explosion-proof antenna 8.

[0038] This invention addresses the problems of traditional firefighting robots, such as limited functionality, cumbersome operation, and lack of perception. Its high-strength protection and mobility modules ensure the vehicle can enter and survive in high-temperature, toxic, and high-collapse-risk core fire zones, replacing firefighters in the most dangerous environments. The highly flexible operation module, with its multiple degrees of freedom and end-effector force sensing capabilities, enables the unmanned vehicle to perform delicate tactical actions previously requiring manual labor, such as opening doors, clearing obstacles, and grasping samples. Simultaneously, force sensors prevent secondary damage to equipment or structures during operation. The omnidirectional environmental perception module constructs a digital twin environment for the vehicle, providing decision-making support for remote operators and laying the foundation for subsequent autonomous driving and intelligent obstacle avoidance.

[0039] During rescue missions, operators remotely control unmanned vehicles to enter the fire scene. The situation is assessed using images and gas data transmitted back by the sensing module. When obstacles need to be addressed, the operator controls the robotic arm's movement; a six-dimensional force sensor provides real-time feedback on contact force information to prevent excessive force; a vision system provides the working perspective. After the task is completed, the robotic arm can... Figure 2 The retractable folding mechanism shown reduces volume for easier transport.

[0040] Example 2: The intelligent control system of an unmanned vehicle, enabling it to operate autonomously and semi-autonomously, describes its process and effects. This system employs a layered architecture of an onboard industrial control computer (upper-level decision-making) and an embedded real-time controller (lower-level execution), executing the following intelligent processes: Data fusion perception: The system first preprocesses multi-source sensor data, including moving average filtering to denoise the six-dimensional force sensor signals and distortion correction of the visual images. Subsequently, based on the weighted Kalman filter algorithm, the system dynamically fuses the processed visual and force data to output key status information such as the distance to environmental obstacles and the magnitude of the contact force between the robotic arm and the object in real time and with high accuracy.

[0041] Intelligent obstacle avoidance decision-making: The system continuously monitors the fused perception information. Once an obstacle is detected at a distance of ≤100cm or the contact force of the robotic arm is ≥ a preset threshold, the obstacle avoidance decision-making algorithm is immediately triggered. Based on real-time perception data from the environmental map, the algorithm automatically generates a new obstacle avoidance path or adjustment instructions, and prioritizes guiding the vehicle or robotic arm to move towards an obstacle-free direction.

[0042] Adaptive Compliant Control: To execute precise obstacle avoidance or operational actions, the system employs a fuzzy adaptive PID control algorithm to drive the robotic arm. This algorithm uses the error *e* between the actual and target postures of the robotic arm and its rate of change *ec* as input. Through a preset 5×5 fuzzy rule matrix, it calculates and adjusts the PID controller parameters (ΔKp, ΔKi, ΔKd) online in real time. These parameters are updated rapidly at a 20ms cycle, driving the servo motors of each joint, enabling the robotic arm's end effector to perform smooth and compliant posture fine-tuning at an extremely low speed of ≤5mm / s.

[0043] See PID control algorithm Figure 7 Specifically: Input and preprocessing: Receive the set value S of the robot arm's attitude angle, process it through the pre-compensation module; acquire the encoder angle signal, process it through the Kalman filter module to obtain the feedback value P.

[0044] Deviation calculation: Calculate the deviation e = S - P between the setpoint S and the feedback value P.

[0045] PID control calculation: Input the deviation e into the PID controller, and perform the following operations: Proportional calculation (Kp×e); Integration operation (Ki×∫edt); Differential operation (Kd×de / dt).

[0046] Control output: Integrating the above calculation results, we obtain the control quantity U = Kp×e + Ki×∫edt + Kd×de / dt.

[0047] Execution and Feedback: The control quantity U is processed by the control limiting module and then drives the servo motor. The motion of the servo motor is collected by the encoder angle sensor and a feedback value P is generated again to complete the closed loop.

[0048] This control system is the core of the unmanned vehicle's intelligence. It upgrades from remote control to autonomous assistance, enables truly precise operation and safe interaction, and ultimately improves the overall mission success rate and reliability.

[0049] The foregoing has provided a detailed description of a multi-degree-of-freedom robotic arm-driven unmanned fire detection vehicle provided by the present invention. Specific examples have been used to illustrate the structure and working principle of the invention. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and core ideas of the present invention. It should be noted that those skilled in the art can make various improvements and modifications to the present invention without departing from its principles, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A fire-fighting and detection unmanned vehicle driven by a multi-degree-of-freedom robotic arm, characterized in that: Comprising The walking chassis module includes a sealed explosion-proof vehicle body (1), a hollow sealed anti-static rubber tire (3) arranged at the lower part of the vehicle body (1), and an embedded lithium battery pack; The multi-degree-of-freedom operation module is arranged at the front end of the upper part of the vehicle body (1) and includes an aluminum alloy double-joint mechanical arm (5) with 10 degrees of freedom, the surface of the mechanical arm (5) is sprayed with a high-temperature-resistant anticorrosive coating, a quick-change clamp (15) is connected to the end of the mechanical arm (5), and a six-dimensional force sensor is installed at the joint of the mechanical arm (5). The environmental perception and communication module includes a multi-in-one gas sensor (2), a laser radar (6), a high-definition explosion-proof pan-tilt camera (7), and an explosion-proof antenna (8) integrated at the rear end of the upper part of the vehicle body (1), and a binocular depth camera (4) arranged at the front end of the vehicle body, which are collectively used for environmental modeling, reconnaissance, and data transmission.

2. The multi-DOF robotic arm driven fire detection and reconnaissance unmanned vehicle according to claim 1, characterized in that: The mechanical arm (5) includes a base fixed to the vehicle body (1), a first joint mechanical arm (20), and a second joint mechanical arm (21); The first joint mechanical arm (20) is pivotally connected to the base through a rotary disc (22) and is driven by a rotary servo motor (10) to realize horizontal rotary motion, and the first joint mechanical arm (20) is driven by a first servo motor (11); The second joint mechanical arm (21) is connected to the end of the first joint mechanical arm (20) and is driven by a second servo motor (12), and the end of the second joint mechanical arm (21) is connected with the clamp (15), and a fixed camera (14) is installed near the clamp.

3. The multi-DOF robotic arm driven fire detection and reconnaissance unmanned vehicle according to claim 1, characterized in that: The clamp (15) includes a mounting base, a pair of mirror-symmetrically arranged bent claws (19), and a pair of connecting pieces, the mounting base is provided with a mounting ear and a fixed shaft, one end of each of the two bent claws (19) is hingedly connected to the mounting ear through a first locking pin (16) and a third locking pin (19), the bent portions of the two bent claws (19) are connected to the connecting pieces, and the other ends of the pair of connecting pieces are hingedly connected to the end of the fixed shaft through a second locking pin (17), thereby forming a parallel four-bar linkage opening and closing mechanism to realize synchronous opening and closing motion of the two claws (19).

4. The multi-DOF robotic arm driven fire detection and reconnaissance unmanned vehicle according to claim 1, characterized in that: The clamp (15) is fixedly connected to the end of the second mechanical arm (21) through the mounting base thereof, and the claws (19) are designed in a modular manner and can be quickly replaced according to different task requirements such as breaking and grabbing.

5. The multi-DOF robotic arm driven fire detection and reconnaissance unmanned vehicle according to claim 1, characterized in that: The vehicle body (1) is a high-strength carbon steel sealed explosion-proof structure capable of resisting 800℃ high temperature, and the wheels (3) are directly driven by explosion-proof motors; the vehicle is equipped with a 200Ah / 48V lithium battery pack supporting ≥4 hours of operation and ≤1.5 hours of fast charging.

6. The multi-DOF robotic arm driven fire detection and reconnaissance unmanned vehicle according to claim 1, characterized in that: Each joint of the mechanical arm (5) is provided with a servo motor and an angle encoder coaxially connected thereto, and the servo motor and the angle encoder constitute a position closed-loop control system, which can drive the end clamp (15) of the mechanical arm (5) to realize 360-degree omnidirectional positioning and continuous attitude adjustment in three-dimensional space.

7. The multi-DOF robotic arm driven fire detection and reconnaissance unmanned vehicle according to claim 6, characterized in that: A first six-dimensional force sensor (9) associated with the first servo motor (11) is arranged at the pivot connection between the first section of the robot arm (20) and the base, and a second six-dimensional force sensor (13) associated with the second servo motor (12) is arranged at the driving joint between the second section of the robot arm (21) and the first section of the robot arm (20).

8. The multi-DOF robotic arm driven fire detection and reconnaissance unmanned vehicle according to claim 1, characterized in that: The six-dimensional force sensor collects X / Y / Z axis force and torque signals in real time, with a measurement accuracy of ≤±0.5% FS, and a force feedback threshold preset in the range of 10-100 N.

9. An intelligent control system for controlling a fire detection unmanned vehicle driven by the multi-degree-of-freedom robot arm according to any one of claims 1 to 8, characterized in that: The control system adopts a layered architecture of an on-board industrial computer and an embedded real-time controller, and performs the following: fusion processing of the six-dimensional force sensor data and the camera visual data to generate environment obstacle distance and robot arm contact force information, autonomous generation of obstacle avoidance control instructions when the obstacle distance ≤100 cm or the contact force ≥ the preset threshold, and driving the robot arm to adjust the posture to achieve obstacle avoidance operation according to the obstacle avoidance instructions.

10. The intelligent control system according to claim 9, characterized in that: The data fusion processing is based on a weighted Kalman filtering algorithm, which dynamically allocates the weights of the visual data and the force sensor data, and outputs key information such as the environment obstacle distance and the contact force after fusion; The driving robot arm to adjust the posture adopts a fuzzy adaptive PID control algorithm; The algorithm takes the posture error (e) and the error change rate (ec) of the robot arm as input, adjusts the PID control parameters online through a 5x5 fuzzy rule matrix, and updates the control instructions at a period of 20 ms; Finally, the robot arm is controlled to adjust the posture at a micro-motion speed of ≤5 mm / s, and the positioning error of the end effector is controlled within ±2 mm.

Images