Autonomous navigation system suitable for complex scene based on dynamic path planning

Abstract

The invention discloses an autonomous navigation system suitable for a complex scene based on dynamic path planning, and relates to the technical field of path planning navigation, the autonomous navigation system comprises a cloud server and a plurality of mobile devices, the cloud server is in wireless connection with each mobile device, and each mobile device is in wireless connection; each mobile device comprises a movable device body, and a sensing module, a decision module, a control module and an interaction module which are arranged on the device body, the cloud server is wirelessly connected with each mobile device, each mobile device is wirelessly connected, the cloud server or one of the mobile devices can be always kept as a host, and the cloud server and one of the mobile devices are wirelessly connected. The method is responsible for dynamic path planning, so that the workload of a main processor in each mobile device is reduced, and the mobile device can quickly respond in a complex scene. The mobile device can cut the surrounding environment into 36 parts through 36 cameras, the collection precision is improved, and the flexibility of image connection between the cameras is effectively compensated for through 5-degree view angle overlapping.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning and navigation, and particularly relates to an autonomous navigation system applicable to complex scenarios based on dynamic path planning. Background Art

[0002] An autonomous navigation system is an intelligent system that integrates technologies such as sensor data, satellite positioning, and inertial measurement unit (IMU) to achieve environmental perception, position positioning, and path planning, and can autonomously execute navigation tasks without manual intervention. The development trends of autonomous navigation systems include improving positioning accuracy and stability, realizing multi-sensor fusion (such as GPS+INS+visual SLAM), and enhancing the ability to adapt to extreme environments (such as heavy fog, dust, etc.).

[0003] However, in the face of complex environments, the autonomous navigation system needs to quickly calculate the moving paths of continuously moving dynamic objects around through the powerful computing power of the current device, and then make path judgment and planning, which brings great pressure to the central CPU of the device. Moreover, when the device is working in a complex environment, it is inevitable that foreign objects will fall and cause the device to collide. If some sensors such as cameras are damaged at this time, it is easy to cause the device to be unable to move normally.

[0004] In summary, an autonomous navigation system applicable to complex scenarios based on dynamic path planning is designed. Summary of the Invention

[0005] In order to overcome the above deficiencies, the present invention provides an autonomous navigation system applicable to complex scenarios based on dynamic path planning.

[0006] The present invention achieves the above object through the following technical solutions:

[0007] An autonomous navigation system applicable to complex scenarios based on dynamic path planning includes a cloud server and a number of mobile devices. The cloud server is wirelessly connected to each mobile device, and each mobile device is wirelessly connected to each other;

[0008] The mobile device includes a movable device body and a perception module, a decision module, a control module, and an interaction module provided on the device body;

[0009] The perception module is used for multi-source information collection of the environment around the mobile device;

[0010] The decision module is used for dynamic path planning according to the situation of the environment around the device;

[0011] The control module is used for receiving the movement instruction of the main processor in the device body and controlling the movement of the device body;

[0012] The interaction module is used for the wireless communication between the device body and other mobile devices or cloud servers, and facilitates on-site interaction operations for staff members.

[0013] Preferably, the sensing module includes a spectral fusion positioning module, an environmental modeling module, and a dynamic monitoring module. The spectral fusion positioning module includes a quantum gyroscope, a photon radar, and a geomagnetic fingerprint, and is used to accurately locate the position of the mobile device, so as to ensure that the cloud server can obtain the accurate position of the mobile device in real time. The environmental modeling module includes 4D laser SLAM and 36 cameras. The cameras are evenly distributed circumferentially around the device body, and there is a 5° angular overlap between the viewing angles of each adjacent camera. It is used to collect the surrounding environment of the mobile device in real time. At the same time, the 36 cameras can divide the surrounding environment into 36 parts, improving the collection accuracy. Moreover, the 5° angular overlap can effectively compensate for the flexibility of image connection between cameras. The dynamic monitoring module includes a millimeter-wave radar array, and the millimeter-wave radars are evenly distributed around the device body. Combined with the cameras, it can monitor dynamic objects around in real time.

[0014] Preferably, the 36 cameras are fixed on a rotating support. When one or more cameras are damaged in a complex environment, the main processor inside the device body will send a rotation instruction to the rotating support, causing the rotating support to rotate periodically to ensure that the remaining cameras can still monitor the surrounding environment of the device body in real time. Each camera is electrically connected to a sub-processor, and each sub-processor is electrically connected to the main processor. The sub-processor processes the information collected by the camera and then transmits the data to the main processor.

[0015] Preferably, the decision-making module includes the following working steps:

[0016] Step 1: Data collection. The main processor collects the data collected by the sensing module.

[0017] Step 2: Determine the host. When the mobile device is wirelessly connected to both the cloud server and other mobile devices, the cloud server serves as the host at this time to perform dynamic path planning, and all the mobile devices connected to the network only need to execute the control operations. When the cloud server loses contact with each mobile device and the mobile devices are wirelessly connected to each other, data is transmitted between them through the interaction module at this time. Then, based on the sensing module, the mobile device located in the middle can be determined, and this mobile device serves as the host, only responsible for executing dynamic path planning. Other mobile devices transmit data to this mobile device and all execute the control operations. When the mobile device loses contact with both the cloud server and other mobile devices, this mobile device serves as the host, performs dynamic path planning based on the collected surrounding environment data, and simultaneously turns on the interaction module to attempt periodic networking with other mobile devices or the cloud server.

[0018] Step 3: When the main processor is a slave, the main processor transmits the data to the host through the interaction module. Then, the main processor receives the dynamic path planning instruction from the host through the interaction module. Next, the main processor controls the movement of the device body through the control module.

[0019] Step 4: When the main processor is the host and is connected to the network, the main processor receives the data from the slave through the interaction module, performs dynamic path planning, and then sends the dynamic path planning instruction to the corresponding slave.

[0020] Step 5: When the main processor is the host and is not connected to the network, the main processor performs dynamic path planning based on the data and controls the movement of the device body through the control module.

[0021] Preferably, when the mobile devices are interconnected and lose contact with the cloud server, the mobile device in the front will also transmit the environmental data to the mobile device in the back through the interaction module, which can reduce the operation of the main processor of the back-end server and improve the efficiency of dynamic path planning.

[0022] The beneficial effects of the present invention are as follows: In the autonomous navigation system applicable to complex scenarios based on dynamic path planning:

[0023] 1. The cloud server is wirelessly connected to each mobile device, and the mobile devices are wirelessly connected, which can always keep the cloud server or one of the mobile devices as the host, responsible for dynamic path planning, thereby reducing the workload of the main processor in each mobile device and enabling the mobile device to respond quickly in complex scenarios.

[0024] 2. The mobile device can divide the surrounding environment into 36 parts through 36 cameras, improving the acquisition accuracy. Moreover, the 5° visual angle overlap can effectively compensate for the flexibility of image connection between cameras.

[0025] 3. Each camera is electrically connected to a sub-processor, and each sub-processor is electrically connected to the main processor. The sub-processor processes the information collected by the camera and then transmits the data to the main processor. In this way, even when the main processor is working alone, the main processor only needs to perform dynamic path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The present invention will be described by way of examples and with reference to the accompanying drawings, wherein:

[0027] Figure 1 is a working step diagram of the decision-making module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] Now, the present invention will be further described in detail with reference to the accompanying drawings. These drawings are all simplified schematic diagrams, only illustrating the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention.

[0029] As Figure 1 shown, an autonomous navigation system applicable to complex scenarios based on dynamic path planning includes a cloud server and a number of mobile devices. The cloud server is wirelessly connected to each mobile device, and each mobile device is wirelessly connected;

[0030] The mobile device includes a movable device body and a sensing module, a decision-making module, a control module, and an interaction module provided on the device body;

[0031] The sensing module is used for multi-source information acquisition of the environment around the mobile device;

[0032] The decision-making module is used for performing dynamic path planning according to the situation of the environment around the device;

[0033] The control module is used for receiving the movement instruction of the main processor in the device body and controlling the movement of the device body;

[0034] The interaction module is used for wireless communication between the device body and other mobile devices or the cloud server, and can facilitate on-site interaction operations by staff.

[0035] Specifically, the perception module includes a spectral fusion positioning module, an environment modeling module, and a dynamic monitoring module. The spectral fusion positioning module includes a quantum gyroscope, a photon radar, and a geomagnetic fingerprint, and is used to accurately locate the position of the mobile device, so as to ensure that the cloud server can obtain the accurate position of the mobile device in real time. The environment modeling module includes a 4D laser SLAM and 36 cameras. The cameras are evenly distributed circumferentially around the device body, and there is a 5° visual angle overlap between the visual angles of each adjacent camera. It is used to collect the surrounding environment of the mobile device in real time. At the same time, the surrounding environment can be cut into 36 parts by the 36 cameras, improving the collection accuracy. Moreover, the 5° visual angle overlap can effectively make up for the flexibility of the image connection between cameras. The dynamic monitoring module includes a millimeter wave radar array. The millimeter wave radars are evenly distributed around the device body. Combined with the cameras, it can monitor the surrounding dynamic objects in real time.

[0036] Specifically, the 36 cameras are fixed on a rotating support. When one or more of the cameras are damaged in a complex environment, the main processor inside the device body will send a rotation instruction to the rotating support, causing the rotating support to rotate periodically to ensure that the remaining cameras can still monitor the surrounding environment of the device body in real time. Each camera is electrically connected to a sub-processor, and each sub-processor is electrically connected to the main processor. The sub-processor processes the information collected by the camera and then transmits the data to the main processor.

[0037] Specifically, the decision-making module includes the following working steps:

[0038] Step 1: Data collection. The main processor collects the data collected by the perception module.

[0039] Step 2: Determine the host. When the mobile device maintains a wireless connection with both the cloud server and other mobile devices, the cloud server serves as the host at this time to perform dynamic path planning, and all connected mobile devices only need to execute control operations; when the cloud server loses contact with each mobile device and the mobile devices maintain a wireless connection with each other, data is transmitted to each other through the interaction module at this time. Then, according to the perception module, the mobile device located in the middle position can be determined, and this mobile device serves as the host, only responsible for executing dynamic path planning, and other mobile devices transmit the data to this mobile device and all execute control operations; when the mobile device loses contact with both the cloud server and other mobile servers, this mobile device serves as the host at this time, and performs dynamic path planning according to the collected surrounding environment data, and at the same time turns on the interaction module to perform periodic networking attempts with other mobile devices or the cloud server.

[0040] Step 3: When the main processor acts as a slave, it transmits data to the host through the interaction module, then receives the dynamic path planning instructions from the host through the interaction module, and then controls the movement of the device body through the control module.

[0041] Step 4: When the main processor acts as the host and is connected to the network, it receives data from the slave through the interaction module, performs dynamic path planning, and then sends the dynamic path planning instructions to the corresponding slave.

[0042] Step 5: When the main processor acts as the host and is not connected to the network, it performs dynamic path planning based on data and controls the movement of the device body through the control module.

[0043] Specifically, when each mobile device is interconnected and loses contact with the cloud server, the mobile device in the front will also transmit environmental data to the mobile device in the rear through the interaction module, thereby reducing the operation of the main processor of the rear server and improving the efficiency of dynamic path planning.

[0044] Case 1: Intelligent factory material handling system

[0045] Scenario requirements: Environment: A 20,000-square-meter workshop with 30 AGVs, mobile robotic arms, and workers working together

[0046] Challenges: High-frequency dynamic obstacles (new obstacles added every 5 seconds), electromagnetic interference (signal attenuation in the welding area > 50%), and path conflicts in multi-device collaboration

[0047] System operation mechanism

[0048] 1. Perception layer collaboration (implemented by hardware)

[0049] Spectral fusion positioning:

[0050] Quantum gyroscope (zero drift error < 0.001° / h) compensates the steering deviation of the AGV in real time;

[0051] Photon radar (wavelength of 1550nm) penetrates the smoke in the welding area to generate a centimeter-level point cloud map;

[0052] Data flow: Each AGV uploads 1.2GB of positioning data to the cloud per second;

[0053] Environmental modeling:

[0054] 36 cameras (Sony IMX789 sensors) achieve 360° rotation per second through rotating mounts to achieve non-blind spot monitoring;

[0055] Fault response: When 3 cameras are damaged by splashing welding slag, the remaining cameras rotate 30° every 2 seconds, with a coverage rate of 97%;

[0056] Dynamic monitoring:

[0057] The millimeter-wave radar array (77 GHz) detects hidden AGVs within 5 m, and the positioning error is <2 cm after fusing with camera data;

[0058] 2. Decision-making layer dynamic programming

[0059] Judge the host, and the host makes dynamic path planning.

[0060] 3. Control and interaction optimization

[0061] Model predictive control (MPC):

[0062] When turning sharply, the torque distribution algorithm reduces the motor response delay from 80 ms to 35 ms

[0063] Energy consumption management:

[0064] Use geomagnetic fingerprints to turn off 50% of the cameras, and the battery life is extended from 12 hours to 18 hours

[0065] Data verification:

[0066] Indicator Standard value Measured value Obstacle avoidance success rate ≥99％ 99.6％ Number of path conflicts ≤1 time / hour 0.3 times / hour Network disconnection recovery time <500ms 218 ms

[0067] Case 2: Hospital surgical instrument delivery robot

[0068] Scene requirements and restrictive conditions: Access permission in the sterile area, prohibited entry into the electromagnetic sensitive area of the MRI room, priority channel for emergency surgeries. Dynamic elements: 60 medical staff movements per hour, channel changes caused by temporary equipment placement. System operation mechanism

[0069] 1. Sensing layer (medical-grade accuracy)

[0070] Cross-floor positioning:

[0071] Quantum gyroscope + photon radar to achieve accurate floor height identification in the elevator shaft (error ±0 floors)

[0072] Geomagnetic fingerprint matching for the position of the metal door in the operating room (positioning error <5 cm)

[0073] Dynamic monitoring:

[0074] Millimeter-wave radar identifies moving hospital beds (triggers emergency avoidance when the speed >1.5 m / s)

[0075] 36 cameras detect dropped objects on the surgical cart through a 5° overlapping view (recognition rate 99.8%)

[0076] 2. Decision-making layer logic (medical compliance)

[0077] Determine the host, and the host performs dynamic path planning.

[0078] 3. Human-computer interaction design

[0079] AR projection interface:

[0080] The light field projection displays the navigation path on the operating room floor (with brightness adaptive adjustment to avoid interfering with doctors), and the bone conduction headphones broadcast voice prompts (sound pressure level < 30 dB, meeting the operating room noise standard)

[0081] Data verification:

[0082] Indicator Standard value Measured value Rate of accidental entry into the aseptic area 0 times / day 0 times / day Emergency avoidance response time <200ms 153 ms MRI room false trigger rate ≤0.1％ 0.05％

[0083] Case 3: Earthquake disaster rescue navigation system

[0084] Scene requirements

[0085] Extreme environment: Building collapse leads to non-line-of-sight environment, damage to communication base stations, and unstable vital signs of survivors

[0086] Dynamic challenges: Aftershocks trigger the risk of secondary structural collapse and high difficulty in coordinating rescue personnel

[0087] System operation mechanism

[0088] 1. Breakthrough in the perception layer (adaptation to extreme environments)

[0089] Non-line-of-sight positioning:

[0090] UWB (3.5 - 6.5 GHz frequency band) penetrates 5 layers of concrete walls and combines with an acoustic wave array to achieve positioning with an accuracy of 0.3 m

[0091] Photon radar scans the collapsed voids to construct a 3D survival space model (modeling speed 10 ㎡ / second)

[0092] Dynamic monitoring:

[0093] Millimeter-wave radar detects the expansion speed of structural cracks (warning threshold > 0.5 mm / s)

[0094] Infrared camera (FLIR Boson 640) identifies the body temperature of survivors (detection distance 15 m)

[0095] 2. Decision-making layer

[0096] Determine the host, and the host performs dynamic path planning.

[0097] 3. Cooperative control and communication

[0098] Self-organizing network communication:

[0099] Adopt the LoRa+Wi-Fi7 hybrid protocol to maintain a communication radius of 500m in a network-disconnected environment

[0100] The front rescue robot transmits environmental data in real time, reducing the computing workload of the rear robot by 67%

[0101] Data verification:

[0102] Indicator Standard value Measured value Non-line-of-sight positioning error <0.5m 0.28m Aftershock early warning lead ≥10 seconds 15 seconds Cooperative path conflict rate ≤2％ 0.8％

[0103] In summary, the system has strong adaptability to dynamic environments. The cloud server and each mobile device will perform rapid and flexible switching according to the network connection situation. Due to the operation of 36 cameras, the workload of the main processor can be reduced, improving the dynamic path planning efficiency and accuracy of the main processor. At the same time, the cameras can be controlled to rotate for image acquisition, enabling effective environmental image acquisition to be maintained in extremely complex environments and improving the reliability of the system in complex environments.

[0104] Based on the inspiration of the present invention, through the above description, relevant staff can make various changes and modifications completely within the scope of not deviating from the technical idea of this invention. The technical scope of this invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. An autonomous navigation system based on dynamic path planning and applicable to complex scenarios, characterized in that: It includes a cloud server and several mobile devices. The cloud server is wirelessly connected to each mobile device, and each mobile device is wirelessly connected to each other. The mobile device includes a movable device body, and a sensing module, a decision-making module, a control module, and an interaction module provided on the device body. The sensing module is used for multi-source information collection of the environment around the mobile device. The decision-making module is used for dynamic path planning according to the situation of the environment around the device. The control module is used to receive the movement instruction of the main processor in the device body and control the movement of the device body. The interaction module is used for wireless communication between the device body and other mobile devices or the cloud server, and can facilitate on-site interaction operations by the staff.

2. The autonomous navigation system applicable to complex scenarios based on dynamic path planning according to claim 1, wherein: The sensing module includes a spectral fusion positioning module, an environment modeling module, and a dynamic monitoring module. The spectral fusion positioning module includes a quantum gyroscope, a photon radar, and a geomagnetic fingerprint. The environment modeling module includes 4D laser SLAM and 36 cameras. The cameras are evenly distributed circumferentially around the device body, and there is a 5° viewing angle overlap between the viewing angles of each adjacent camera. The dynamic monitoring module includes a millimeter-wave radar array, and the millimeter-wave radars are evenly distributed around the device body.

3. The autonomous navigation system applicable to complex scenarios based on dynamic path planning according to claim 2, characterized in that: The 36 cameras are fixed on a rotating support, and each camera is electrically connected to a secondary processor, and each secondary processor is electrically connected to the main processor.

4. The autonomous navigation system applicable to complex scenarios based on dynamic path planning according to claim 1, characterized in that: The decision-making module includes the following working steps: Step 1: Data collection, where the main processor collects the data collected by the sensing module. Step 2: Determine the host. When the mobile device maintains wireless connections with both the cloud server and other mobile devices, the cloud server serves as the host at this time to perform dynamic path planning, and all the mobile devices in the network only need to execute the control operation. When the cloud server loses contact with each mobile device and the mobile devices maintain wireless connections with each other, data is transmitted to each other through the interaction module at this time, and then according to the sensing module, the mobile device located in the middle position can be judged, and this mobile device serves as the host, only responsible for executing dynamic path planning, and other mobile devices transmit the data to this mobile device and all execute the control operation. When the mobile device loses contact with both the cloud server and other mobile servers, this mobile device serves as the host at this time, and performs dynamic path planning according to the collected surrounding environment data, and at the same time turns on the interaction module to perform periodic networking attempts with other mobile devices or the cloud server. Step 3: When the main processor is a slave, the main processor transmits the data to the host through the interaction module, then the main processor receives the dynamic path planning instruction from the host through the interaction module, and then the main processor controls the movement of the device body through the control module. Step 4: When the main processor is the host and is connected to the network, the main processor receives the data from the slave through the interaction module, and performs dynamic path planning, and then sends the dynamic path planning instruction to the corresponding slave. Step 5: When the main processor is the host and is not connected to the network, the main processor performs dynamic path planning through the data, and controls the movement of the device body through the control module.

5. The autonomous navigation system applicable to complex scenarios based on dynamic path planning according to claim 1, wherein: When the mobile devices are interconnected and lose contact with the cloud server, the mobile device in the front will still transmit environmental data to the mobile device in the rear through the interaction module.

Images