Intelligent blood sampling robot system based on space positioning

Abstract

The invention relates to the technical field of medical logistics, and discloses an intelligent blood sampling robot system based on spatial positioning, which comprises a robot body, a central controller, a spatial positioning module, a dynamic environment modeling module, a motion control module, a transportation path planning module and a man-machine interaction module. According to the invention, the flight traffic mode is combined with the optimal path flight, the limitation of ground traffic congestion is eliminated, the transportation time is shortened, and in the path planning, the task continuity in a complex environment is ensured based on the synchronous positioning and map construction algorithm and the ultra wide band wireless carrier technology and in combination with the three-dimensional space coordinate sequence and the dynamic path adjustment, and the path planning efficiency is improved. The method has high dynamic response adjustment capability and timeliness, and a high-timeliness and high-safety intelligent solution is provided for emergency blood transfusion, cross-regional blood product blending and other scenes.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of medical logistics, and particularly relates to an intelligent blood taking robot system based on spatial positioning. BACKGROUND

[0002] The current medical logistics field is faced with core pain points such as low efficiency of traditional transportation, poor adaptability to dynamic environment and insufficient safety. The blood distribution mode mainly based on ground transportation is difficult to meet the stringent requirements of timeliness in emergency scenes due to the limitations of urban traffic congestion, road planning and high labor costs.

[0003] For example, in some megacities, the time consumption of traditional land transportation of blood is generally 60-90 minutes, and in extreme cases, it may take more than 2 hours, while the unmanned aerial vehicle for blood delivery has realized efficient transportation of 15-50 minutes. However, the existing medical robots for medical logistics mostly rely on fixed paths or manual scheduling, lack real-time response capability to complex environments (such as sudden obstacles and weather changes), and the transportation path is easy to be tampered with or hijacked, which is difficult to guarantee the safety of medical supplies. SUMMARY

[0004] The application aims to provide an intelligent blood taking robot system based on spatial positioning to solve the technical problems in the background art.

[0005] To achieve the above-mentioned purpose, the application discloses the following technical scheme: an intelligent blood taking robot system based on spatial positioning, comprising: a robot body, the inside of the robot body is provided with a shockproof cargo cabin, the top of the robot body is provided with a flight execution part, and the front of the robot body is provided with a switchable cabin door corresponding to the shockproof cargo cabin; a central controller arranged on the robot body; a spatial positioning module arranged on the robot body and in communication connection with the central controller, configured to perform centimeter-level spatial positioning in a preset virtual channel through a positioning sensor of the flight execution part and a ground communication base station based on a simultaneous localization and mapping algorithm and an ultra-wideband wireless carrier technology; a dynamic environment modeling module embedded in the central controller and configured to construct a building structure and an obstacle point cloud map through a scanning result of a laser radar, and generate a virtual path of a three-dimensional spatial coordinate sequence based on an AI algorithm; a motion control module embedded in the central controller and configured to be in control connection with the flight execution part; The transportation path planning module embedded in the central controller is configured to store preset transportation routes and positioning data fed back by the spatial positioning module and virtual paths fed back by the dynamic environment modeling module, dynamically adjust pre-selected transportation routes, and encrypt key nodes in the paths; The human-computer interaction module embedded in the central controller is configured to control opening and closing of the switchable cabin door based on human-computer interaction.

[0006] Preferably, the spatial positioning module comprises: The multi-sensor fusion positioning unit is configured to perform centimeter-level dynamic positioning through multi-channel signal interaction between the positioning sensor of the flight execution member and the ground communication base station based on a simultaneous localization and mapping algorithm, an ultra-wideband wireless carrier technology, and an inertial navigation system. The anomaly correction unit is configured to trigger a BeiDou grid code-based auxiliary positioning algorithm to correct spatial coordinate errors when positioning data deviation exceeds a preset positioning threshold.

[0007] Preferably, the motion control module comprises: The PID control unit is configured to dynamically adjust PID control parameters to stabilize the attitude of the robot body according to real-time state data of the flight execution member and virtual path parameters. The emergency obstacle avoidance unit is configured to preferentially execute an emergency obstacle avoidance instruction based on vector field navigation when detecting that the distance to an obstacle is less than a preset safety threshold, and simultaneously feed back path conflict information to the transportation path planning module.

[0008] Preferably, the auxiliary positioning algorithm comprises: The fusion weight of the BeiDou grid code and the ultra-wideband wireless carrier signal is dynamically adjusted according to environmental noise intensity and signal-to-noise ratio of the positioning sensor. The grid coordinates of the BeiDou grid code and the time difference positioning data of the ultra-wideband wireless carrier signal are weighted and fused based on the adjusted fusion weight to generate corrected spatial coordinates. The corrected spatial coordinates are smoothed by a Kalman filtering algorithm, and the final spatial coordinates are output.

[0009] Preferably, the dynamic environment modeling module comprises: The point cloud data processing unit is configured to generate a three-dimensional point cloud map of building structures and obstacles through multi-angle scanning results of the laser radar, and perform semantic segmentation on the point cloud data based on a graph neural network to identify passable areas and dangerous areas. The virtual path generation unit is configured to generate a virtual path of three-dimensional spatial coordinate sequences based on a deep reinforcement learning algorithm and a generative adversarial network. The virtual path optimization unit is configured to dynamically adjust the virtual path of the three-dimensional spatial coordinate sequence according to real-time environmental changes based on a reinforcement learning algorithm, and generate an obstacle avoidance strategy library for calling by the transportation path planning module.

[0010] Preferably, the transportation path planning module comprises: The path storage unit is configured to store a preset transportation route. The multi-objective optimization unit is configured to generate an optimal path based on the selected transportation route, positioning data fed back by the spatial positioning module in real time, and the virtual path fed back by the dynamic environment modeling module, by a genetic algorithm, wherein the genetic algorithm is executed to optimize the objectives including flight time, energy consumption coefficient, and obstacle avoidance probability. The path verification unit is configured to perform hash encryption on key nodes of the optimal path by a blockchain technology, and share encrypted path data with a ground communication base station.

[0011] Preferably, the virtual path optimization unit is further configured to: The virtual path information is shared with adjacent robots to cooperatively plan a group flight path.

[0012] Preferably, the human-computer interaction module comprises: The biometric verification unit is configured to verify the legality of blood taking and placing operations of a user by any one or more of a plurality of verification mechanisms including fingerprint recognition, iris recognition, and voice recognition, and only open the opening and closing authority of the switchable cabin door after the legality verification is passed. The remote control unit is configured to display the flight state of the robot body through a remote display interface, and provide a manual takeover flight control function in an emergency.

[0013] Preferably, the remote control unit is further configured to: The virtual trajectory of the robot is superimposed on the actual environment in real time through an AR device to assist an operator in path correction and emergency intervention.

[0014] Preferably, the biometric verification unit is further configured to: When the legality verification fails, an audible and visual alarm is triggered, and alarm information is sent to the starting point and the ending point at the same time.

[0015] Beneficial effects: the intelligent blood taking robot system based on spatial positioning of the application adopts the traffic mode of flight combined with optimal path flight, gets rid of the restriction of ground traffic congestion, shortens the transportation time, and in the path planning, based on the simultaneous localization and mapping algorithm and the ultra-wideband wireless carrier technology, combined with the three-dimensional spatial coordinate sequence and dynamic path adjustment, ensures the task continuity in complex environment, has high dynamic response adjustment ability and timeliness, and provides a high timeliness and high safety intelligent solution for emergency blood transfusion, cross-regional blood product allocation and other scenes. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0017] Figure 1 The structural block diagram of the intelligent blood taking robot system based on spatial positioning provided by the embodiments of the present application is shown in the figure. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely below. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] In this paper, the term "including" is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitation, the elements defined by the statement "including" do not exclude the presence of other identical elements in the process, method, article or equipment including the elements.

[0020] For the traditional unmanned aerial vehicle or robot system, it is difficult to achieve high-precision positioning in complex buildings or indoor environments, which relies on a single sensor (such as GPS or inertial navigation) for positioning; environmental modeling often uses static maps, which cannot adapt to real-time obstacle changes; the path planning algorithm is mostly preset route, lacking dynamic adjustment ability to sudden conditions. The embodiments provide a high timeliness and high safety aerial solution for medical logistics, suitable for emergency blood transfusion, cross-regional blood product allocation and other scenes, and help to build low-altitude life route network in cities.

[0021] By the above, the embodiment provides an intelligent blood taking robot system based on spatial positioning as shown in the drawings, comprising a robot body, a central controller, a spatial positioning module, a dynamic environment modeling module, a motion control module, a transportation path planning module and a human-computer interaction module. Figure 1

[0022] In detail The robot body adopts a lightweight composite material frame (such as a carbon fiber frame), and an anti-vibration cargo cabin is arranged inside. A temperature control system (such as a Peltier cooling plate) is arranged in the cabin to maintain the constant temperature (4-8℃) of blood or other blood products. A flight execution component (such as a four-rotor flight assembly) is arranged on the top, and a switchable cabin door corresponding to the anti-vibration cargo cabin is arranged on the front. The switchable cabin door is linked and controlled to open and close by an electromagnetic lock and a biological recognition verification unit described below.

[0023] The central controller is installed on the robot body. The central controller can be based on an ARM Cortex-A9 processor, integrated with a multi-core computing unit, responsible for coordinating data interaction and real-time decision-making of each module.

[0024] The spatial positioning module is installed on the robot body and communicates with the central controller. Based on a simultaneous localization and mapping algorithm, a local map is constructed, combined with an ultra-wideband wireless carrier technology, and through the positioning sensor of the flight execution component and the ground communication base station, a centimeter-level spatial positioning in a preset virtual channel is performed.

[0025] Further, the spatial positioning module comprises: A multi-sensor fusion positioning unit is configured to perform centimeter-level dynamic positioning through multi-channel signal interaction between the positioning sensor of the flight execution component and the ground communication base station based on a simultaneous localization and mapping algorithm, an ultra-wideband wireless carrier technology and an inertial navigation system. Specifically, the simultaneous localization and mapping algorithm constructs a local map through laser radar point cloud matching and ultra-wideband wireless carrier (UWB) distance measurement. The inertial navigation system (INS) provides short-term inertial data through a gyroscope and an accelerometer. The fusion algorithm adopts Kalman filtering, dynamically adjusts the weight of UWB and INS according to the signal strength of the sensor, for example, increases the weight of UWB when the signal is strong, and increases the compensation ratio of INS when the signal is weak. An abnormal correction unit is configured to trigger a Beidou grid code-based auxiliary positioning algorithm to correct spatial coordinate errors when the positioning data deviation exceeds a preset positioning threshold. The coordinate error is corrected by the geographic grid code provided by the Beidou satellite. The correction process dynamically adjusts the correction amplitude in combination with the environmental noise intensity.

[0026] Feasibly, the auxiliary positioning algorithm comprises: ​According to the environmental noise intensity and the positioning sensor signal-to-noise ratio, the fusion weight of the Beidou grid code and the ultra-wideband wireless carrier signal is dynamically adjusted to improve the positioning robustness in complex scenarios, for example, reducing the UWB weight in a high-noise environment and improving the compensation of the INS; Based on the adjusted fusion weight, the grid coordinates of the Beidou grid code and the time difference positioning data of the ultra-wideband wireless carrier signal are weighted and fused to generate corrected spatial coordinates; The corrected spatial coordinates are smoothed by using a Kalman filter algorithm to eliminate instantaneous noise interference, and the final spatial coordinates are output.

[0027] The dynamic environment modeling module is embedded in the central controller, scans the building structure and obstacles by laser radar 360° scanning, constructs a point cloud map of the building structure and obstacles, and generates a virtual path of a three-dimensional spatial coordinate sequence based on an AI algorithm.

[0028] Further, the dynamic environment modeling module comprises: The point cloud data processing unit is configured to: the laser radar scans the environment at a frequency of 10Hz, generates a three-dimensional point cloud map of the building structure and obstacles through a graph neural network (GNN) model based on the multi-angle scanning results of the laser radar, and performs semantic segmentation on the point cloud data based on the graph neural network to identify passable areas and dangerous areas; The virtual path generation unit is configured to generate a virtual path of a three-dimensional spatial coordinate sequence based on a deep reinforcement learning algorithm and a generative adversarial network; The virtual path optimization unit is configured to dynamically adjust the virtual path of the three-dimensional spatial coordinate sequence, i.e., dynamically adjust the generation strategy of the virtual track, based on a reinforcement learning algorithm, and generate an obstacle avoidance strategy library for the transportation path planning module to call.

[0029] Feasibly, the generation steps of the virtual path are as follows: Environment modeling and semantic encoding: use a graph neural network (GNN) to model the graph structure of the point cloud data, extract environmental topological features (such as road width, obstacle density, building height), and generate a labeled three-dimensional semantic map combined with the semantic segmentation results; Initial track generation: based on the generator module of the generative adversarial network, input the three-dimensional semantic map and the preset transportation task parameters (such as starting point, ending point, maximum flight height, obstacle avoidance safety distance), output the rough path of the candidate virtual path; the generator module adopts the architecture of combining a multilayer perceptron with a 3D convolutional neural network, learns the mapping relationship from environmental features to track paths by training historical environment-track data sets; Reinforcement Learning Optimization: Input the candidate trajectory into a deep reinforcement learning model, and perform multiple rounds of iterative optimization through the state space (current environmental features, robot position, obstacle distribution), action space (trajectory adjustment direction and amplitude), and reward function (obstacle avoidance success rate, path smoothness, energy consumption coefficient) to generate a virtual path that meets safety and efficiency constraints; Dynamic Adaptive Adjustment: During flight, real-time centimeter-level positioning data from the spatial positioning module and the latest point cloud map from the dynamic environment modeling module are received, and the virtual path is dynamically corrected through online learning mechanisms (such as incremental reinforcement learning) to ensure consistency with real-time environmental changes.

[0030] The motion control module is embedded in the central controller and connected to the flight execution component control to control the flight of the flight execution component.

[0031] Further, the motion control module includes: A PID control unit configured to dynamically adjust PID control parameters (such as higher speed, larger proportional coefficient; larger attitude angle fluctuation, smaller integral coefficient) based on real-time state data of the flight execution component (such as speed, attitude angle, angular velocity, etc.) and virtual path parameters to stabilize the attitude of the robot body during flight; An emergency obstacle avoidance unit configured to prioritize executing emergency obstacle avoidance instructions based on vector field navigation when detecting that the distance to an obstacle is less than a preset safety threshold, and the robot adjusts the path in the direction away from the obstacle and simultaneously feeds back path conflict information to the transportation path planning module.

[0032] The transportation path planning module is embedded in the central controller and is used to store preset transportation routes and positioning data fed back by the spatial positioning module and virtual paths fed back by the dynamic environment modeling module, dynamically adjust the pre-selected transportation routes, and encrypt key nodes (such as starting point and ending point coordinates) in the path.

[0033] Further, the transportation path planning module includes: A path storage unit configured to store preset transportation routes; A multi-objective optimization unit configured to generate an optimal path based on the selected transportation route, real-time positioning data fed back by the spatial positioning module, and virtual paths fed back by the dynamic environment modeling module through a genetic algorithm. The genetic algorithm optimizes the objectives including flight time, energy consumption coefficient, and obstacle avoidance probability when executed, i.e., the genetic algorithm evaluates the objectives of flight time, energy consumption, and obstacle avoidance probability to select the optimal path combination through simulating the natural selection process; A path verification unit is configured to hash encrypt key nodes (such as starting point and ending point coordinates) of the optimal path by using a blockchain technology, and share encrypted path data with a ground communication base station, so as to prevent the path from being tampered with or hijacked.

[0034] It is feasible that the virtual path optimization unit is further configured to: The virtual path information is shared with adjacent robots to cooperatively plan a group flight path, so as to avoid flight conflicts. The cooperative planning can be realized by a distributed decision mechanism. The robots detect path conflicts of adjacent robots, and adjust flight height or delay take-off time to avoid collision.

[0035] A human-computer interaction module is embedded in the central controller, and is used to control the switch of the switchable cabin door based on human-computer interaction, so as to ensure the legality of blood taking and placing operation.

[0036] Further, the human-computer interaction module comprises: A biometric verification unit is configured to verify the legality of blood taking and placing operation of a user by using any one or more of a fingerprint recognition, an iris recognition and a voice recognition verification mechanism, and to open the switch permission of the switchable cabin door only after the legality verification is passed; wherein the fingerprint recognition collects feature points by using a capacitive sensor, the iris recognition extracts texture features by using a near-infrared camera, and the voice recognition is matched based on voiceprint features. In the application, the corresponding receiving person (the operator of blood taking or placing) of the starting point and the ending point first enters fingerprint, iris and / or voiceprint data as a sample in the verification mechanism, and then performs single or multiple verification by using any one or more of the fingerprint recognition, the iris recognition and the voice recognition. If the receiving person is not the designated receiving person of the current transportation, the verification will fail (the fingerprint, iris and / or voiceprint data is not entered in advance and / or is not selected as the receiving person corresponding to the current transportation). A remote control unit is configured to transmit state information of the robot by using a 5G network, display the flight state of the robot body to a remote end (such as an administrator or an operator) through a remote display interface, and provide a manual takeover flight control function in an emergency.

[0037] It is feasible that the remote control unit is further configured to: The virtual trajectory of the robot is superimposed on the actual environment in real time by using an AR device, so as to assist an operator in path correction and emergency intervention. When the operator wears AR glasses, the virtual trajectory is superimposed on the actual environment. Based on a pre-set instruction, path correction is realized by triggering an emergency control instruction, such as an “urgent stop” by gesture recognition (such as “clenching a fist”).

[0038] In addition, it is feasible that the biometric verification unit is further configured to: When the legality verification fails, an audible and visual alarm is triggered and an alarm message is sent to the starting point and the ending point (e.g., the operator and the receiver corresponding to the starting point, the operator and the receiver corresponding to the ending point, etc.).

[0039] In summary, the intelligent blood collection robot system based on spatial positioning in the embodiment can fly along the optimal path in three-dimensional space, avoiding ground traffic congestion, and combining laser radar and AI algorithms to achieve centimeter-level obstacle avoidance and shorten transportation time. Through multi-sensor fusion positioning and reinforcement learning algorithms, the robot can adjust the virtual track in real time to deal with temporary obstacles or weather changes (such as strong winds and moderate rain), ensuring the continuity of transportation tasks. Based on graph neural networks and generative adversarial networks, the virtual track of the three-dimensional spatial coordinate sequence is dynamically generated, and a multi-objective optimization algorithm is used to balance flight time, energy consumption, and obstacle avoidance probability. The path verification unit uses blockchain technology to hash encrypt key nodes and share data with ground base stations to prevent malicious tampering of the path and ensure traceability throughout the transportation process. Adjacent robots share virtual track information and use swarm intelligence algorithms to avoid air collisions, improving transportation density and safety in low-altitude economic scenarios. Operators can use the AR interface of the 5G network to correct the flight trajectory in real time and use fingerprint / iris / voice triple verification to ensure the legality of blood collection and placement operations, reducing the risk of human error. Based on this, the core pain points of poor timeliness, weak dynamic response capability, and insufficient safety in traditional medical logistics are addressed, providing an intelligent solution with high timeliness, high safety, and low energy consumption for emergency blood transfusion and cross-regional blood product allocation scenarios, and promoting the construction of urban low-altitude life route networks.

[0040] In the embodiments provided by the present application, it should be understood that the embodiments described herein can be realized by hardware, software, firmware, middleware, codes or any proper combination thereof. For hardware implementation, the processor can be realized in one or more of the following components: an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, other electronic units designed to perform the functions described herein, or a combination thereof. For software implementation, the procedures described herein can be implemented with a computer program that directs relevant hardware to complete the procedures. When implemented, the above program can be stored in a computer readable storage medium or transmitted as one or more instructions or codes on a computer readable storage medium. The computer readable storage medium includes computer storage media and communication media including any medium that facilitates the transfer of a computer program from one place to another. The storage medium can be any available medium that can be accessed by a computer. The computer readable storage medium can include but not limited to RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage medium or other magnetic storage devices, or any other medium that can be used to carry or store desired program codes in the form of instructions or data structures and can be accessed by a computer.

[0041] Finally, it should be noted that the above only refers to the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, modifications or equivalent replacements of the technical solutions described in the foregoing embodiments can be made by those skilled in the art, or some technical features can be replaced by equivalent features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A spatially-positioning based intelligent blood taking robot system, characterized in that, The application relates to a robot body, a central controller, a space positioning module, a dynamic environment modeling module, a motion control module, a transportation path planning module and a man-machine interaction module. The robot body is internally provided with a shockproof load cabin, a top part is provided with a flight execution part, and a front part is provided with a switchable cabin door corresponding to the shockproof load cabin. The central controller is arranged on the robot body. The space positioning module is arranged on the robot body, is in communication connection with the central controller, and is configured to perform centimeter-level space positioning in a preset virtual channel through a positioning sensor of the flight execution part and a ground communication base station based on a simultaneous localization and mapping algorithm and an ultra-wideband wireless carrier technology. The dynamic environment modeling module is embedded in the central controller and is configured to construct a building structure and an obstacle point cloud map through scanning results of a laser radar and generate a virtual path of a three-dimensional space coordinate sequence based on an AI algorithm. The motion control module is embedded in the central controller and is configured to be in control connection with the flight execution part. The transportation path planning module is embedded in the central controller and is configured to store a preset transportation route, positioning data fed back by the space positioning module and a virtual path fed back by the dynamic environment modeling module, dynamically adjust a preselected transportation route and encrypt key nodes in the path. The man-machine interaction module is embedded in the central controller and is configured to control opening and closing of the switchable cabin door based on man-machine interaction.

2. The spatially-positioning based intelligent blood taking robot system according to claim 1, wherein, The space positioning module comprises a multi-sensor fusion positioning unit configured to perform centimeter-level dynamic positioning through multi-channel signal interaction between the positioning sensor of the flight execution part and the ground communication base station based on a simultaneous localization and mapping algorithm, an ultra-wideband wireless carrier technology and an inertial navigation system. An abnormality correction unit is configured to trigger a Beidou grid code-based auxiliary positioning algorithm to correct space coordinate errors when positioning data deviation exceeds a preset positioning threshold. The motion control module comprises a PID control unit configured to dynamically adjust PID control parameters to stabilize the attitude of the robot body according to real-time state data of the flight execution part and virtual path parameters.

3. The spatially-positioning based intelligent blood taking robot system according to claim 2, wherein, An emergency obstacle avoidance unit is configured to preferentially execute an emergency obstacle avoidance instruction based on vector field navigation when detecting that the distance of an obstacle is less than a preset safety threshold and synchronously feed back path conflict information to the transportation path planning module. The auxiliary positioning algorithm comprises dynamically adjusting the fusion weight of the Beidou grid code and the ultra-wideband wireless carrier signal according to environmental noise intensity and signal-to-noise ratio of the positioning sensor. Based on the adjusted fusion weight, the grid coordinates of the Beidou grid code and the time difference positioning data of the ultra-wideband wireless carrier signal are weighted and fused to generate corrected space coordinates.

4. The spatially-positioned based intelligent blood taking robot system according to claim 3, wherein, The corrected space coordinates are smoothed by using a Kalman filtering algorithm, and the final space coordinates are output. The dynamic environment modeling module comprises a point cloud data processing unit configured to generate a three-dimensional point cloud map of a building structure and an obstacle through multi-angle scanning results of a laser radar and perform semantic segmentation on point cloud data based on a graph neural network to identify passable areas and dangerous areas. ​ ​ 5. The spatially-positioning based intelligent blood taking robot system according to claim 1, wherein, ​ ​ The virtual path generation unit is configured to generate a virtual path of a three-dimensional coordinate sequence based on a deep reinforcement learning algorithm and a generative adversarial network; The virtual path optimization unit is configured to dynamically adjust the virtual path of the three-dimensional coordinate sequence according to real-time environmental changes based on a reinforcement learning algorithm, and generate an obstacle avoidance strategy library for calling by the transport path planning module.

6. The spatially-positioned based intelligent blood taking robot system according to claim 5, wherein, The transport path planning module comprises: A path storage unit configured to store a preset transport route; A multi-objective optimization unit configured to generate an optimal path based on the selected transport route, positioning data fed back by the spatial positioning module in real time, and the virtual path fed back by the dynamic environment modeling module, by a genetic algorithm, wherein the genetic algorithm is executed to optimize the objectives including flight time, energy consumption coefficient, and obstacle avoidance probability; A path verification unit configured to perform hash encryption on key nodes of the optimal path by using a blockchain technology, and share encrypted path data with a ground communication base station.

7. The spatially-positioned based intelligent blood taking robot system according to claim 5, wherein, The virtual path optimization unit is further configured to: Perform collaborative planning of a group flight path by sharing virtual path information with adjacent robots.

8. The spatially-positioned based intelligent blood taking robot system according to any one of claims 3 or 6, wherein, The human-computer interaction module comprises: A biometric verification unit configured to verify the legality of blood taking and placing operations of a user by using any one or more of a plurality of verification mechanisms including fingerprint recognition, iris recognition, and voice recognition, and to open the opening and closing permission of the switchable cabin door only after the legality verification is passed; A remote control unit configured to display the flight state of the robot body through a remote display interface, and to provide a manual takeover flight control function in an emergency.

9. The spatially-positioned based intelligent blood taking robot system of claim 8, wherein, The remote control unit is further configured to: Overlay the virtual trajectory of the robot and the actual environment in real time through an AR device to assist an operator in path correction and emergency intervention.

10. The spatially-positioned based intelligent blood taking robot system of claim 8, wherein, The biometric verification unit is further configured to: Trigger an audible and light alarm and send alarm information to the starting point and the ending point simultaneously when the legality verification fails.

Images