A guide cane with a six-legged traction mechanism and an active traction obstacle avoidance method

Abstract

This invention discloses a guide cane with a six-legged traction mechanism and an active traction obstacle avoidance method. The guide cane includes a hollow tube, a handle, a control box, a six-legged traction component, a telescopic drive component, and a control component. The bottom end of the hollow tube is open, and the six-legged traction component is located in the open portion of the hollow tube. The telescopic drive component is located inside the hollow tube, enabling two actions: retracting the six-legged traction component into the open portion and removing it from the open portion. By installing a camera in the control box at the upper part of the hollow tube, this invention can acquire more comprehensive environmental information and effectively eliminate noise and false alarms in the visual data through a vision processor. It can identify common obstacles such as pedestrians, vehicles, steps, and utility poles, improving the accuracy of obstacle avoidance judgment and solving the safety hazards of traditional guide canes that can only perceive obstacles through physical contact and have severely delayed obstacle avoidance.

Description

Technical Field

[0001] This invention relates to the field of guide tools, and in particular to a guide cane with a six-legged traction mechanism and an active traction and obstacle avoidance method. Background Technology

[0002] Guide canes are an important assistive tool to help visually impaired people travel. Traditional guide canes only perceive obstacles on the ground through physical contact, which is a single function and cannot identify obstacles in the air or at a distance in advance, resulting in poor safety.

[0003] With the development of smart technology, electronic guide canes with obstacle perception functions, whose information collection modules are deployed at the end of the cane, have emerged. However, the following problems still exist: First, the information collection module is easily obstructed by ground debris due to height limitations, making it difficult to obtain sufficient environmental information and leading to errors in obstacle avoidance judgment. Second, after the guide cane transmits obstacle avoidance information to the visually impaired person, the user needs to actively adjust the direction of travel according to the prompts, which can easily lead to collisions or getting lost, especially in complex road conditions. Third, although some guide devices have attempted to introduce mobile mechanisms, they suffer from poor portability, making it difficult to flexibly retract and extend, and inconvenient for daily carrying and use. Summary of the Invention

[0004] To address one or more of the aforementioned problems, the present invention provides a guide cane with a six-legged traction mechanism and an active traction obstacle avoidance method.

[0005] According to one aspect of the present invention, a guide cane with a six-legged traction mechanism includes a hollow tube, a handle, a control box, a six-legged traction assembly, a telescopic drive assembly, and a control assembly.

[0006] The bottom end of the hollow tube is open, and the six-legged traction assembly is set in the open part of the hollow tube. The telescopic drive assembly is set inside the hollow tube, and the telescopic drive assembly realizes two actions: the six-legged traction assembly is stored in the open part and the six-legged traction assembly is moved out of the open part.

[0007] The handle is fixedly connected to the top of the hollow tube, and interactive components are provided on the handle.

[0008] The control box is fixedly connected to the upper middle part of the hollow tube. A camera is installed on the control box, and a vision processor and a control module are installed inside the control box. The vision processor is electrically connected to the camera, and the control module is electrically connected to the six-legged traction component, the telescopic drive component, the interaction component, the camera, and the vision processor. The control module is used to receive environmental information collected by the vision processor and control the six-legged traction component to apply traction force to achieve active traction guidance.

[0009] In some embodiments, the hexapod traction assembly includes a base on which six robotic arm components arranged in a circumferential array are mounted on the upper surface, and anti-slip blocks are fixedly connected to the bottom surface of the base.

[0010] When the six-legged traction assembly is in the extended state, the anti-slip block is removed from the ground, and the six robotic arm components are in contact with the ground.

[0011] When the six-legged traction assembly is in the retracted state, the six robotic arm components move upwards and are inside the hollow tube, with the bottom surface of the anti-slip block resting against the ground.

[0012] In some embodiments, the robotic arm component includes a second motor mounted on a base, a first swing arm mounted on the output shaft of the second motor, a third motor mounted on the first swing arm, a second swing arm mounted on the output shaft of the third motor, a fourth motor mounted on the second swing arm, and a robotic gripper mounted on the output shaft of the fourth motor.

[0013] In some embodiments, a control button is installed on the upper surface of the hollow tube and is electrically connected to the first motor. A GPS locator is installed inside the hollow tube and is electrically connected to the control module.

[0014] In some implementations, the interaction components include two vibration sensors and a voice sensor, with the two vibration sensors mounted on opposite sides of the handle and the voice sensor mounted inside the control box.

[0015] In some implementations, the interaction components include two vibration sensors and a voice sensor, with the two vibration sensors mounted on opposite sides of the handle and the voice sensor mounted inside the control box.

[0016] In some implementations, a GPS locator is installed inside the hollow tube, and the GPS locator is electrically connected to the control module.

[0017] An active traction and obstacle avoidance method for a guide cane with a six-legged traction mechanism includes the following steps: S1. Acquire images and depth information of the foreground environment using a camera. Then, use a sliding extreme value averaging filter algorithm to denoise the N consecutive frames of sampled data. The filter output formula is: ; in, For the ki-th sampled data, These represent the maximum and minimum values ​​within the sliding window, respectively, and N is the window length, with a default value of 10. S2. The vision processor uses the YOLOv8 algorithm to perform target detection on the preprocessed image, outputting the category, coordinates, and distance information of obstacles. Its total loss function is: ; in, For CIoU bounding box regression loss, For classifying losses, For distribution focus loss; S3. When an obstacle is detected and the distance is less than the safety threshold. At that time, an integral angle planning algorithm is used to generate local obstacle avoidance trajectories. Specifically: Phase 1: Obstacle avoidance trigger phase. The fuselage yaw angle increases linearly with the distance traveled, as shown in the formula: ; in, This is the yaw direction coefficient. The maximum obstacle avoidance yaw angle is s, and the cumulative forward distance is s. To smoothly adjust the distance; Phase 2: Obstacle Avoidance Maintenance Phase, when And when the obstacle is still within a safe distance, maintain the yaw angle unchanged; Phase 3: Correction Phase, when the distance to the obstacle is greater than the distance away threshold. At that time, the fuselage yaw angle decreases linearly with time integral, as shown by the formula: ; in, This refers to the cumulative time during the recovery phase. S4. Based on the planned foot target coordinates P(x,y,z), solve for the joint angles of a single 3-DOF robotic arm. The inverse solution formula is: ; in, The length of the base segment. The length of the thigh link. The length of the lower leg connecting rod. The lateral swing angle, The hip joint angle, The knee joint angle; S5. Using the control module, the joint angle command obtained from the solution is sent to the six-legged traction component, driving the six-legged traction component to move according to the planned trajectory, applying a proportional traction force in the avoidance direction, as shown in the formula: ; Where k is the proportionality coefficient, with a value ranging from 0.5 to 2 N / m. This represents the distance between the current position and the target trajectory. S6. Obstacle avoidance information is synchronously fed back through the interactive components on the handle. Users can switch between active traction mode and ordinary cane mode with one click by extending the lifting pole.

[0018] In some implementations, in step S3, the rotational transformation formula for converting the fuselage yaw angle into the coordinates of the target at each leg end is: ; in, Let be the initial foot coordinates of the i-th leg. It represents the forward linear velocity.

[0019] In some implementations, in step S4, the foot end of the swing phase adopts a cosine lifting trajectory, and the height changes with time according to the following formula: ; in, To normalize time, For the duration of the oscillation phase, This is the maximum lifting height.

[0020] In some implementations, in step S5, the hexapod traction assembly moves using a triangular gait, a quadrupedal gait, or a undulating gait. Triangular gait: The six legs are divided into group A (LF, RM, LH) and group B (RF, LM, RH). The two groups alternate between support and swing, with a period of T=1.0-1.5s and a support ratio β=0.6-0.7. Quadrupedal gait: The six legs are divided into three pairs, each pair swinging synchronously in turn, with all four legs supporting the body at all times, and the support ratio is 0.7-0.8; Wavy gait: Six legs swing in sequence, with only one leg swinging at a time and the other five legs supporting the body, with a support ratio β=0.83.

[0021] The beneficial effects of this guide cane with a six-legged traction mechanism and its active traction obstacle avoidance method are as follows: First, by installing a camera in the control box located in the upper part of the hollow tube, more comprehensive environmental information can be obtained. Furthermore, the vision processor effectively eliminates noise and false alarms in the visual data, identifying common obstacles such as pedestrians, vehicles, steps, and utility poles, thus improving the accuracy of obstacle avoidance judgment and solving the safety hazard of traditional guide canes that can only perceive obstacles through physical contact and suffer from severe delays in obstacle avoidance. Second, through the coordinated design of the six-legged traction component and the control module, the control module directly drives the six-legged robotic arm to apply traction force in the avoidance direction after receiving environmental information, without... The system allows users to independently determine the direction of travel and adjust their posture to achieve active traction guidance. Thirdly, an interactive component composed of vibration and voice sensors provides intuitive guidance on avoidance direction through vibrations at different locations, while simultaneously broadcasting the type and distance of obstacles via voice, further improving obstacle avoidance response speed. Fourthly, by setting up a telescopic drive component, pressing the control button will extend the six-legged traction component when active traction is needed, entering active traction mode. When active traction is not needed, pressing the button again will completely retract the six-legged robotic arm into the hollow tube, and the anti-slip block will contact the ground, switching the device to ordinary cane mode for easy daily carrying and storage. Attached Figure Description

[0022] Figure 1 This is a three-dimensional schematic diagram (a) of a guide cane with a six-legged traction mechanism according to an embodiment of the present invention. Figure 2 for Figure 1 A cross-sectional view of the guide cane for the visually impaired. Figure 3 for Figure 1 The diagram shows a three-dimensional illustration of the handle mounted on the hollow tube. Figure 4 for Figure 1 The diagram shows the structure of the six-legged traction assembly in its open state. Figure 5 for Figure 1 The diagram shows the structure of the six-legged traction assembly in its stowed state. Figure 6 for Figure 1 The diagram shows the structure of the six-legged traction assembly. Hollow tube 1, handle 2, control box 3, camera 4, interactive components 5; The six-legged traction assembly 6, base 61, mechanical claw 62, anti-slip block 63, second motor 64, first swing arm 65, third motor 66, second swing arm 67, and fourth motor 68; Telescopic drive assembly 7, fixed base 71, lifting rod 72, first motor 73, gear 74, rack 75, support plate 76, control button 77. Detailed Implementation

[0023] The invention will now be described in further detail with reference to the accompanying drawings. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, while the terms "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0024] Figures 1 to 6 A guide cane with a six-legged traction mechanism according to an embodiment of the present invention is schematically shown. As shown, the device includes a hollow tube 1, a handle 2, a control box 3, a six-legged traction assembly 6, a telescopic drive assembly 7, and a control assembly.

[0025] The bottom end of the hollow tube 1 is open. The six-legged traction assembly 6 is located at the open part of the hollow tube 1. The telescopic drive assembly 7 is located inside the hollow tube 1. The telescopic drive assembly 7 realizes two actions: the six-legged traction assembly 6 is stored in the open part and the six-legged traction assembly 6 is moved out of the open part.

[0026] The open design at the bottom of the hollow tube 1 provides storage space for the six-legged traction assembly 6, allowing for the concealed storage and retraction of the telescopic drive assembly 7 and the six-legged traction assembly 6, meeting daily carrying needs.

[0027] The handle 2 is fixedly connected to the top of the hollow tube 1, and the handle 2 is equipped with an interactive component 5.

[0028] The control box 3 is fixedly connected to the upper middle part of the hollow tube 1. A camera 4 is installed on the control box 3. A vision processor and a control module are installed inside the control box 3. The vision processor is electrically connected to the camera 4. The control module is electrically connected to the six-legged traction component 6, the telescopic drive component 7, the interaction component 5, the camera 4, and the vision processor. The control module is used to receive environmental information collected by the vision processor and control the six-legged traction component 6 to apply traction force to achieve active traction guidance.

[0029] The control box 3 is installed in the upper part of the hollow tube 1, so that the camera 4 is in a high position for acquisition, which effectively avoids obstruction by ground debris and obtains a wider field of view and long-distance environmental information.

[0030] Preferably, the six-legged traction assembly 6 includes a base 61, on the upper surface of which six robotic arm components are mounted in a circumferential array, and on the bottom surface of the base 61, an anti-slip block 63 is fixedly connected.

[0031] The six robotic arm components are arranged in a circular array to ensure uniform force distribution during traction and improve walking stability. The anti-slip block 63 at the bottom of the base 61 is made of rubber, which increases the friction with the ground when the anti-slip block 63 contacts the ground, improving the anti-slip performance in ordinary cane mode.

[0032] When the six-legged traction assembly 6 is in the extended state, the anti-slip block 63 is detached from the ground, and the six robotic arm components are in contact with the ground.

[0033] When extended, the anti-slip block 63 detaches from the ground to avoid motion interference with the robotic arm components, ensuring that the robotic arm components can freely perform traction actions and provide stable active traction force.

[0034] When the six-legged traction assembly 6 is in the retracted state, the six robotic arm components move upward and are inside the hollow tube 1, and the bottom surface of the anti-slip block 63 rests against the ground.

[0035] In the retracted state, the robotic arm components are completely housed inside the hollow tube 1. The device is the same size as a regular cane, making it easy to carry and store. Furthermore, the anti-slip block 63 contacts the ground, allowing the device to switch to the regular cane mode to meet the needs of different scenarios.

[0036] The robotic arm component includes a second motor 64 mounted on a base 61, a first swing arm 65 mounted on the output shaft of the second motor 64, a third motor 66 mounted on the first swing arm 65, a second swing arm 67 mounted on the output shaft of the third motor 66, a fourth motor 68 mounted on the second swing arm 67, and a robotic gripper 62 mounted on the output shaft of the fourth motor 68.

[0037] The first swing arm 65 is driven to swing by the output shaft of the second motor 64, which in turn causes the mechanical gripper 62 to swing. The second swing arm 67 is driven to swing by the output shaft of the third motor 66, which in turn causes the mechanical gripper 62 to swing. The mechanical gripper 62 is directly driven to swing by the output shaft of the third motor 66, thus achieving the effect of multi-degree-of-freedom adjustment of the mechanical gripper 62.

[0038] Furthermore, the telescopic drive assembly 7 includes a fixed base 71 and a support plate 76 fixedly connected inside the hollow tube 1. A first motor 73 is mounted on the fixed base 71, and a gear 74 is mounted on the output shaft of the first motor 73. A lifting rod 72 is slidably connected to the support plate 76. A rack 75 is mounted on the lifting rod 72, and the rack 75 meshes with the gear 74. One end of the lifting rod 72 is fixedly connected to the base 61.

[0039] The transmission method using gears 74 and racks 75 can precisely control the extension and retraction distance of the six-legged traction assembly 6. The support plate 76 provides support and guidance for the lifting rod 72, preventing swaying during lifting and improving structural stability.

[0040] Preferably, a control button 77 is installed on the upper surface of the hollow tube 1, and the control button 77 is electrically connected to the first motor 73.

[0041] When the user holds the handle 2, they can easily touch the control button 77 to control the extension and retraction of the six-legged traction component 6 with one button.

[0042] Furthermore, the interaction component 5 includes two vibration sensors and a voice sensor. The two vibration sensors are respectively installed on both sides of the handle 2, and the voice sensor is installed inside the control box 3.

[0043] The vibration sensors on both sides provide intuitive guidance on the avoidance direction through vibrations at different locations, eliminating the need for users to rely on auditory judgment and improving obstacle avoidance response speed. Meanwhile, the voice sensor enables voice command input and environmental information broadcasting, enriching human-computer interaction methods and enhancing the user experience.

[0044] Preferably, a GPS locator is installed inside the hollow tube 1, and the GPS locator is electrically connected to the control module.

[0045] GPS locators can obtain the user's location information in real time and, combined with the control module, enable route navigation, helping visually impaired people plan the best travel route and avoid getting lost.

[0046] Working principle: When the user needs to use the active traction function, the user presses the control button 77, and the control module 32 sends a forward rotation command to the motor 73. The motor 73 drives the gear 74 to rotate, and the gear 74 drives the rack 75 and the lifting rod 72 to move downward, pushing the six-legged traction assembly 6 out of the opening of the hollow tube 1. After the six-legged traction assembly 6 is fully extended, the control module controls the six-legged robotic arm 62 to open to the working posture, the motor 73 stops rotating and locks, and the equipment enters the active traction mode.

[0047] At this time, camera 4 collects RGB images and depth information within a range of 1-10m in front and transmits them to the vision processor. The vision processor uses a sliding extreme value removal average filtering algorithm to denoise 10 consecutive frames of sampled data, and then uses the YOLOV8 algorithm to identify obstacles, pedestrians, traffic signs and other information in the environment. The processed results are then transmitted to the control module.

[0048] The control module combines the navigation data from the GPS locator to generate a global travel route. When an obstacle is detected to be less than the safety threshold, an integral angle planning algorithm is used to generate a local obstacle avoidance trajectory. The inverse kinematics algorithm is used to solve the angles of each joint of the six-legged robotic arm 62 and then sends control commands to the six-legged traction component 6.

[0049] At this moment, the six-legged robotic arm 62 moves along the planned trajectory, applying a slight traction force in the avoidance direction, directly causing the user to change the direction of travel.

[0050] During active traction, the control module activates the vibration sensor on the corresponding side based on the obstacle's location, and simultaneously broadcasts the obstacle type, distance, and avoidance direction via the voice sensor, providing multi-dimensional feedback to the user.

[0051] When the user needs to use a regular cane, the user presses the control button 77 again. The control module controls the six-legged robotic arm 62 to retract upwards, and then sends a reverse command to the motor 73. The motor 73 drives the gear 74 to reverse, which drives the rack 75 and the lifting rod 72 to move upwards, and stores the six-legged traction assembly 6 inside the hollow tube 1.

[0052] Once the six-legged traction component 6 is fully retracted, the motor 73 locks, the anti-slip block 63 contacts the ground, and the device switches to ordinary cane mode.

[0053] An active traction and obstacle avoidance method for a guide cane with a six-legged traction mechanism includes the following steps: S1. Acquire images and depth information of the foreground environment through camera 4, and perform noise reduction processing on N consecutive frames of sampled data using a sliding extreme value averaging filter algorithm. The filter output formula is: ; in, For the ki-th sampled data, These represent the maximum and minimum values ​​within the sliding window, respectively, and N is the window length, with a default value of 10. S2. The vision processor uses the YOLOv8 algorithm to perform target detection on the preprocessed image, outputting the category, coordinates, and distance information of obstacles. Its total loss function is: ; in, For CIoU bounding box regression loss, For classifying losses, For distribution focus loss; S3. When an obstacle is detected and the distance is less than the safety threshold. At that time, an integral angle planning algorithm is used to generate local obstacle avoidance trajectories. Specifically: Phase 1: Obstacle avoidance trigger phase. The fuselage yaw angle increases linearly with the distance traveled, as shown in the formula: ; in, This is the yaw direction coefficient. The maximum obstacle avoidance yaw angle is s, and the cumulative forward distance is s. To smoothly adjust the distance; Phase 2: Obstacle Avoidance Maintenance Phase, when And when the obstacle is still within a safe distance, maintain the yaw angle unchanged; Phase 3: Correction Phase, when the distance to the obstacle is greater than the distance away threshold. At that time, the fuselage yaw angle decreases linearly with time integral, as shown by the formula: ; S4. Based on the planned foot target coordinates P(x,y,z), solve for the joint angles of a single 3-DOF robotic arm. The inverse solution formula is: ; in, The length of the base segment. The length of the thigh link. The length of the lower leg connecting rod. The lateral swing angle, The hip joint angle, The knee joint angle; S5. Using the control module, the joint angle command obtained from the solution is sent to the six-legged traction component, driving the six-legged traction component to move according to the planned trajectory, applying a proportional traction force in the avoidance direction, as shown in the formula: ; Where k is the proportionality coefficient, with a value ranging from 0.5 to 2 N / m. This represents the distance between the current position and the target trajectory. S6. Obstacle avoidance information is synchronously fed back through the interactive component 5 on the handle 2. Users can switch between active traction mode and ordinary cane mode with one click by extending the lifting rod 72.

[0054] In some implementations, in step S3, the rotational transformation formula for converting the fuselage yaw angle into the coordinates of the target at each leg end is: ; in, Let be the initial foot coordinates of the i-th leg. It represents the forward linear velocity.

[0055] In some implementations, in step S4, the foot end of the swing phase adopts a cosine lifting trajectory, and the height changes with time according to the following formula: ; in, To normalize time, For the duration of the oscillation phase, This is the maximum lifting height.

[0056] In some embodiments, in step S5, the hexapod traction assembly 6 moves in a triangular gait, a quadrupedal gait, or a undulating gait. Triangular gait: The six legs are divided into group A (LF, RM, LH) and group B (RF, LM, RH). The two groups alternate between support and swing, with a period of T = 1.0-1.5s and a support ratio β = 0.6-0.7. Quadrupedal gait: The six legs are divided into three pairs, each pair swinging synchronously in turn, with all four legs supporting the body at all times, and the support ratio is 0.7-0.8; Wavy gait: Six legs swing in sequence, with only one leg swinging at a time and the other five legs supporting the body, with a support ratio β=0.83.

[0057] The active traction obstacle avoidance process of this application is as follows: First, the user holds the handle 2, and the camera 4 captures RGB images and depth information within a range of 1-10m in front of them, which are then transmitted to the vision processor.

[0058] Then, the user presses the control button 77 on the handle 2 to issue a six-legged extension command. The control module sends a control signal to the motor 73 and the six-legged traction assembly 6. The motor 73 rotates forward, and the gear 74 drives the lifting rod 72 to move down inside the hollow tube 1 through the rack 75, pushing the six-legged traction assembly 6 out of the hollow tube 1. After it is fully extended, the six-legged robotic arm 62 opens to the working posture.

[0059] Subsequently, the vision processor performs a sliding extreme value removal average filter on the 10 consecutive frames of sampled data. The filter output formula is as follows: ; By removing noise and false positives from the data, stable environmental information is obtained.

[0060] Next, the vision processor uses the YOLOv8 algorithm to perform object detection on the preprocessed image, with the total loss function being: ; It identifies common obstacles such as pedestrians, vehicles, steps, and utility poles, outputs the category, coordinates, and distance information of the obstacles, and transmits the data to the control module.

[0061] When control module 4 detects that the distance to the obstacle is less than the safety threshold When the distance reaches 2m, the obstacle avoidance procedure is activated.

[0062] If obstacle avoidance is triggered: Determine the obstacle's location; if the obstacle is on the right... The fuselage veered to the left; The yaw angle increases linearly with the distance traveled, as shown in the formula: ; in, .

[0063] If obstacle avoidance is maintained: when If the distance to the obstacle is still less than 2m, maintain the yaw angle and continue moving forward.

[0064] If the distance to the obstacle is greater than the distance away threshold When the yaw rate is 3m, the fuselage yaw angle decreases linearly with time, as shown in the formula: ; Until they return to their original direction of travel.

[0065] At this point, the rotational transformation formula for converting the fuselage yaw angle into the coordinates of the targets at the ends of each leg is: ; Assume the link length of a single robotic arm is Based on the target coordinates P(x,y,z) at the foot end, the angles of each joint are solved using inverse kinematics formulas: ; The swing phase foot end adopts a cosine lifting trajectory, and the height formula is: This ensures that the feet are lifted and landed smoothly.

[0066] Subsequently, the joint angle command obtained by the solution is sent to the six-legged traction component 6 through the control module, driving the six-legged traction component 6 to move according to the planned trajectory and apply a proportional traction force in the avoidance direction, as shown in the formula: ; The maximum traction force should not exceed 10N to avoid user discomfort.

[0067] Then, the vibration sensors on both sides of the grip 2 prompt the user to move to the left, while the voice module broadcasts a warning message to the user.

[0068] Finally, after the user reaches the destination, press the control button 77 again. The control module controls the six-legged robotic arm 62 to retract, the motor 73 reverses, and drives the lifting rod 72 to move upward, so that the six-legged traction component 6 is put into the hollow tube 1. The anti-slip block 63 contacts the ground, and the device switches to ordinary cane mode.

[0069] The above descriptions are merely some embodiments of the present invention. Those skilled in the art can make various modifications and improvements without departing from the inventive concept of the present invention, and these all fall within the scope of protection of the present invention.

Claims

1. A blind guiding and obstacle avoiding stick with six-legged traction mechanism, characterized in that, It includes a hollow tube (1), a handle (2), a control box (3), a six-legged traction assembly (6), a telescopic drive assembly (7), and a control assembly; The bottom end of the hollow tube (1) is open, the six-legged traction assembly (6) is located at the open part of the hollow tube (1), and the telescopic drive assembly (7) is located inside the hollow tube (1). The telescopic drive assembly (7) realizes two actions: the six-legged traction assembly (6) is stored in the open part and the six-legged traction assembly (6) is moved out of the open part. The handle (2) is fixedly connected to the top of the hollow tube (1), and an interactive component (5) is provided on the handle (2). The control box (3) is fixedly connected to the upper middle part of the hollow tube (1). A camera (4) is installed on the control box (3). A vision processor and a control module are installed inside the control box (3). The vision processor is electrically connected to the camera (4). The control module is electrically connected to the hexapod traction assembly (6), the telescopic drive assembly (7), the interaction assembly (5), the camera (4), and the vision processor. The control module is used to receive environmental information collected by the vision processor and control the hexapod traction assembly (6) to apply traction force to achieve active traction guidance.

2. The blind guiding and obstacle avoiding stick with hexapod traction mechanism according to claim 1, characterized in that, The six-legged traction assembly (6) includes a base (61), on the upper surface of which are mounted six robotic arm components arranged in a circular array, and on the bottom surface of which are fixedly connected anti-slip blocks (63). When the six-legged traction assembly (6) is in the extended state, the anti-slip block (63) is detached from the ground, and the six robotic arm components are in contact with the ground; When the six-legged traction assembly (6) is in the retracted state, the six robotic arm components move upward and are inside the hollow tube (1), and the bottom surface of the anti-slip block (63) rests against the ground.

3. The blind guiding and obstacle avoiding stick with six-legged traction mechanism according to claim 2, characterized in that, The robotic arm component includes a second motor (64) mounted on a base (61), a first swing arm (65) mounted on the output shaft of the second motor (64), a third motor (66) mounted on the first swing arm (65), a second swing arm (67) mounted on the output shaft of the third motor (66), a fourth motor (68) mounted on the second swing arm (67), and a robotic gripper (62) mounted on the output shaft of the fourth motor (68).

4. The blind guiding and obstacle avoiding stick with hexapod traction mechanism according to claim 2, characterized in that, The telescopic drive assembly (7) includes a fixed base (71) and a support plate (76) fixedly connected inside the hollow tube (1). A first motor (73) is installed on the fixed base (71). A gear (74) is installed on the output shaft of the first motor (73). A lifting rod (72) is slidably connected on the support plate (76). A rack (75) is installed on the lifting rod (72), and the rack (75) meshes with the gear (74). One end of the lifting rod (72) is fixedly connected to the base (61).

5. The blind guiding and obstacle avoiding stick with hexapod traction mechanism according to claim 4, characterized in that, A control button (77) is installed on the upper surface of the hollow tube (1), and the control button (77) is electrically connected to the first motor (73). A GPS locator is installed inside the hollow tube (1), and the GPS locator is electrically connected to the control module.

6. The guide cane with a six-legged traction mechanism according to claim 1, characterized in that, The interactive component (5) includes two vibration sensors and a voice sensor. The two vibration sensors are respectively installed on the two sides of the handle (2), and the voice sensor is installed in the control box (3).

7. An active traction and obstacle avoidance method for a guide cane with a six-legged traction mechanism, comprising the guide cane with a six-legged traction mechanism as described in claim 1, characterized in that, Includes the following steps: S1. Acquire images and depth information of the environment in front of the camera (4), and use the sliding extreme value averaging filter algorithm to denoise the N consecutive frames of sampled data. The filter output formula is: ； in, For the ki-th sampled data, These represent the maximum and minimum values ​​within the sliding window, respectively, and N is the window length, with a default value of 10. S2. The vision processor uses the YOLOv8 algorithm to perform target detection on the preprocessed image, outputting the category, coordinates, and distance information of obstacles. Its total loss function is: ； in, For CIoU bounding box regression loss, For classifying losses, For distribution focus loss; S3. When an obstacle is detected and the distance is less than the safety threshold. At that time, an integral angle planning algorithm is used to generate local obstacle avoidance trajectories. Specifically: Phase 1: Obstacle avoidance trigger phase. The fuselage yaw angle increases linearly with the distance traveled, as shown in the formula: ； in, This is the yaw direction coefficient. The maximum obstacle avoidance yaw angle is s, and the cumulative forward distance is s. To smoothly adjust the distance; Phase 2: Obstacle Avoidance Maintenance Phase, when And when the obstacle is still within a safe distance, maintain the yaw angle unchanged; Phase 3: Correction Phase, when the distance to the obstacle is greater than the distance away threshold. At that time, the fuselage yaw angle decreases linearly with time integral, as shown by the formula: ； in, This refers to the cumulative time during the recovery phase. S4. Based on the planned foot target coordinates P(x,y,z), solve for the joint angles of a single 3-DOF robotic arm. The inverse solution formula is: ； in, The length of the base segment. The length of the thigh link. The length of the lower leg connecting rod. The lateral swing angle, The hip joint angle, The knee joint angle; S5. Using the control module, the joint angle command obtained from the solution is sent to the six-legged traction component (6), driving the six-legged traction component (6) to move according to the planned trajectory, and applying a proportional traction force in the avoidance direction, the formula of which is: ； Where k is the proportionality coefficient, with a value ranging from 0.5 to 2 N / m. This represents the distance between the current position and the target trajectory. S6. Obstacle avoidance information is synchronously fed back through the interactive component (5) on the handle (2). Users can switch between active traction mode and ordinary cane mode with one click by extending the lifting rod (72).

8. The active traction obstacle avoidance method according to claim 7, characterized in that, In step S3, the rotation transformation formula for converting the fuselage yaw angle into the coordinates of the target at each leg end is as follows: ； in, Let be the initial foot coordinates of the i-th leg. It represents the forward linear velocity.

9. The active traction obstacle avoidance method according to claim 7, characterized in that, In step S4, the foot end of the swing phase adopts a cosine lifting trajectory, and the height changes with time according to the following formula: ； in, To normalize time, For the duration of the oscillation phase, This is the maximum lifting height.

10. The active traction obstacle avoidance method according to claim 7, characterized in that, In step S5, the hexapod traction assembly (6) moves in a triangular gait, a quadrupedal gait, or a wavering gait: Triangular gait: The six legs are divided into group A (LF, RM, LH) and group B (RF, LM, RH). The two groups alternate between support and swing, with a period of T=1.0-1.5s and a support ratio β=0.6-0.

7. Quadrupedal gait: The six legs are divided into three pairs, each pair swinging synchronously in turn, with all four legs supporting the body at all times, and the support ratio is 0.7-0.8; Wavy gait: Six legs swing in sequence, with only one leg swinging at a time and the other five legs supporting the body, with a support ratio β=0.83.

Images