All-terrain environment adaptive intelligent following robot

Abstract

The invention relates to an all-terrain environment adaptive intelligent following robot which comprises a followed person gait detection mechanism and an intelligent following robot. The foot end position and the posture of the followed person are used as the expected foot end position; the vertical acceleration is calculated according to the height difference of the sequence foot end positions of the followed person; the torque of all joints is deduced reversely according to the horizontal acceleration calculated in the starting, accelerating, decelerating and stopping states and the weight of the robot, and moving and following of the intelligent following robot are achieved. The visual analysis and radar wave ranging module using the high-definition camera as input is only used for discovering a newly inserted obstacle. The navigation mode with the foot end of the followed person as guidance avoids road surface modeling, planning and obstacle avoidance algorithms under complex road conditions, gravity center control is combined, the intelligent following robot can stably follow and pass through complex sections such as rails and riverbeds which are difficult to pass through by traditional wheel type and multi-legged robots in an all-terrain environment in a load-bearing mode, and the intelligent following robot has high practical value. The method has obvious technical advantages and practical value.

Description

Technical Field

[0001] This invention relates to the field of intelligent following robot technology, specifically to an all-terrain environment-adaptive intelligent following robot. Background Technology

[0002] In scenarios such as outdoor hiking, border patrols, emergency rescue and disaster relief, field logistics, large-scale warehousing, and outdoor event support, traditional transportation tools are limited by damaged roads, narrow spaces, or complex terrain, making it difficult to achieve rapid and accurate delivery of supplies. All-terrain adaptable intelligent following robots, by continuously following and carrying loads on sand, gravel, snow, stairs, mud, and 45° steep slopes, can achieve "fully equipped" delivery even when a single person or team is "empty-handed," and have a wide range of applications.

[0003] However, existing technologies, such as multimodal chassis (wheel-track-leg composite structure) and active suspension adaptive adjustment, and UWB + vision + LiDAR fusion positioning, are still insufficient to truly achieve all-terrain environment adaptation. Specifically, they struggle to identify potential obstacles and plan reasonable paths, such as deep pits covered in grass or dry riverbeds; they have difficulty navigating complex road conditions, such as railway tracks, roads covered with large gravel, mud, or steep slopes, as attempting to traverse them may result in rollover risks; and they struggle to operate stably in tunnels, forests, or indoor environments without GPS. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a technical solution for an all-terrain environment-adaptive intelligent following robot.

[0005] This all-terrain adaptive intelligent following robot includes a person being followed, a gait detection mechanism for the person being followed, and the main body of the intelligent following robot.

[0006] The person being followed wears a gait detection device and walks naturally, actively choosing a suitable path, avoiding obstacles, and stepping on relatively flat, stable, safe, and foot-accommodating locations. Ideally, the heel of the person being followed has a clear following mark; Ideally, the person being followed should use a probe stick to detect hidden road conditions ahead, such as road conditions under grass or under roads covered by water, and find a relatively flat, stable, safe place to stand before walking safely. Preferably, the gait detection mechanism of the person being followed is used to detect and transmit the geographical location of the person's feet in real time, including horizontal and vertical positions (three-dimensional position information of x, y, z axes), the angle between the foot and the horizontal section, and the force distribution on the foot. Preferably, the gait detection mechanism for the followed person includes a positioning and attitude measurement unit, a pressure sensor, an image tracking marker, a laser tracking marker, a laser receiving sensor, and a communication device. The positioning and attitude measurement unit measures the position information of the geometric centers of the forefoot and heel of the followed person, including horizontal and vertical positions. The pressure sensor measures the pressure exerted on the forefoot and heel of the followed person. The image tracking marker is used by the intelligent following robot for identification and positioning via a binocular vision sensor to correct accumulated positioning deviations from the positioning and attitude measurement unit. The laser tracking marker and the laser receiving sensor are used to receive the identification from the intelligent following robot and perform positioning by emitting laser signals from a laser emitter to correct accumulated positioning deviations from the positioning and attitude measurement unit. The communication module reads the latest position and pressure information of the followed person's feet in real time and sends this information to the main body of the intelligent following robot in real time.

[0007] The intelligent following robot moves by completely following footprints. This includes receiving the geographical location of the followed person's feet from a gait detection mechanism, including horizontal and vertical displacements relative to the starting position, and the angle between the foot and the horizontal plane as the desired foot position. The desired displacement is the difference between the robot's current foot position and the desired foot position. The desired total acceleration is calculated by detecting changes in height based on the followed person's current foot position and the adjacent foot position in front, and by calculating horizontal acceleration based on the state of movement (starting or stopping). Combining this with its own weight and the force distribution received from the followed person's feet, the robot determines the force control applied to various joint positions, including the force applied to the coronal plane abduction / adduction joint of the robot's thigh relative to the hip (referred to as the hip joint in this invention). The system enables the intelligent following robot to overcome its own gravitational acceleration while obtaining the desired horizontal and vertical acceleration. The sagittal plane flexion / extension joint of the thigh relative to the hip (referred to as the hip joint in this invention) enables the intelligent following robot to overcome its own gravitational acceleration while obtaining the desired horizontal and vertical acceleration. The sagittal plane flexion / extension joint of the lower leg relative to the thigh (referred to as the knee joint in this invention) enables the intelligent following robot to overcome its own gravitational acceleration while obtaining the desired horizontal and vertical acceleration and the desired force distribution on the foot.

[0008] Preferably, the main body of the intelligent following robot includes five modules: a main controller, a luggage rack controller, a walking planning device, a walking execution device, and an auxiliary device.

[0009] The main controller is used to coordinate and control all electronic components inside the main body of the intelligent following robot.

[0010] Preferably, the main controller includes a communication module, a walking control module, and a center of gravity control module. The communication module receives gait-related sensor data from the gait detection mechanism of the person being followed. The walking control module combines local sensor data from the walking planning device to complete walking planning and executes walking through the walking execution device. The center of gravity control module controls the center of gravity during walking to achieve stable walking.

[0011] The aforementioned walking planning device completes the footprint planning for the intelligent following robot to fully follow the footprints of the person being followed through route planning, thereby avoiding complex road surface modeling, route planning and obstacle avoidance algorithms.

[0012] The walking planning device receives the three-dimensional positions of the x, y, and z axes measured by the positioning and attitude measurement units on the feet of the person being followed from the gait detection mechanism sent by the communication module in the main controller. It compares these positions with the actual three-dimensional positions of its own feet on the x, y, and z axes. Combined with the distance and position information of the following marker on the rear foot of the person being followed measured by the laser emission / receiver sensor or binocular vision sensor, it performs real-time correction and adjustment, and calculates the desired foot trajectory.

[0013] Preferably, the walking planning device includes a positioning and attitude measurement unit, a high-definition camera, a video analyzer, a laser emitter, a binocular vision sensor, a foot pressure sensor, and a radar ranging module. The positioning and attitude measurement unit is installed on the foot of the intelligent following robot and is used to measure the current actual position of the robot's foot. The difference between the current position and the expected position of the followed person's foot received through the main controller communication module is the expected displacement. The high-definition camera is installed on the head of the intelligent following robot to achieve human or canine biological visual biomimicry and capture real-time video of the intelligent following robot's walking direction. The video analyzer analyzes the video stream captured by the high-definition camera in real time to locate the heel of the followed person. Following the marker, the laser emitter continuously emits laser signals to the laser receiver of the gait detection mechanism for distance measurement and positioning. This corrects the accumulated deviation between the positioning and attitude measurement units in the walking planning device and the positioning and attitude measurement units in the gait detection mechanism of the followed person, thereby improving the following accuracy. The binocular vision sensor provides another way to eliminate the accumulated measurement errors of the positioning and attitude measurement units through visual distance measurement and positioning, so as to maximize the following accuracy. The foot pressure sensor is used to control the foot landing and control the center of gravity before possible missteps or sinking, preventing the intelligent following robot from falling. The high-definition camera, video analysis, and radar ranging module work together to detect possible obstacles that may be inserted into the intelligent following robot's path in real time and provide real-time reminders.

[0014] Preferably, the positioning and attitude measurement unit measures the foot position of the intelligent following robot body, measures its actual position, calculates the deviation from the desired position, and then continuously corrects it through the walking execution device to get infinitely closer to the desired position.

[0015] The walking actuator is used to enable the intelligent following robot to walk and completely follow the geographical location of the followed person's feet, including the horizontal and vertical positions in the x and y axis directions.

[0016] Preferably, the walking actuator includes a rigid link of the robot leg, a robot leg joint, a motor located at the joint, and motor control, etc. The rigid link of the robot leg is composed of physical rods, and the robot leg joint is a movable mechanical component connecting adjacent rigid links of the robot leg. The motor located at the joint controls the robot leg joint, including the robot hip joint, hip joint, knee joint, and ankle joint. The motor control realizes force control of the torque of the joint.

[0017] Preferably, the foot of the intelligent following robot and the foot of the person being followed have the same or substantially the same geometry and size as human feet.

[0018] Preferably, the bottom of the intelligent following robot's foot and the sole of the person being followed have the same or substantially the same deformation force and friction force.

[0019] The luggage rack controller is used to control the center of gravity in various gaits during walking, so as to achieve stable walking.

[0020] The luggage rack controller includes luggage rack rails, a luggage tray, actuator motors, and a status register. The center of gravity control module in the main controller calculates the desired center of gravity based on the current gait and compares it with the current center of gravity position stored in the status register. Within the intelligent following robot's body, such as its back, it moves the luggage tray along the luggage rack rails, and through the actuator motors, moves it up, down, left, and right. This modifies the center of gravity of the luggage carried by the intelligent following robot relative to its body, thus achieving balance during weight-bearing walking.

[0021] The auxiliary device receives voice commands from the person being followed and executes commands such as start and stop.

[0022] The auxiliary device includes a microphone and an audio analysis module. The microphone receives voice input commands from the person being followed, such as commands to "pause" or "follow". The audio analysis module analyzes the audio stream of the microphone in real time, extracts the voice that matches the voiceprint information of the person being followed, and identifies the content of the voice. If commands such as "pause" or "follow" are detected, the walking execution device is notified.

[0023] This invention employs the technical solution of an all-terrain adaptive intelligent following robot. The desired foot position is determined by the foot position and posture of the followed person. Vertical acceleration is calculated from the height difference of the followed person's sequential foot positions, and horizontal acceleration is calculated from the start, acceleration, deceleration, and stop states. Combined with the robot's own weight, the torque of all joints, including the hip, hip, knee, and ankle joints, is derived in reverse. In this invention, the visual analysis technology using a high-definition camera as input and the radar ranging module are only used to detect newly inserted obstacles on the desired travel route. The basic technical principle of this invention, and the foot-guided navigation mode, avoids the need for road surface modeling, route planning, and obstacle avoidance algorithms in complex road conditions. Combined with center of gravity control, this enables the intelligent following robot to stably follow under load in all-terrain environments, traversing complex terrains that are difficult for traditional wheeled and multi-legged robots to cross, such as steps, railway tracks, and shallow water-covered riverbeds. This demonstrates significant technical advantages and practical value. Attached Figure Description

[0024] Figure 1 This is a schematic diagram illustrating the working principle of the all-terrain environment-adaptive intelligent following robot of the present invention.

[0025] Figure 2 This is a schematic diagram of the gait detection mechanism for the person being followed according to the present invention.

[0026] Figure 3 This is a schematic diagram of the internal structure of the all-terrain environment-adaptive intelligent following robot of the present invention.

[0027] Figure 4 This is a schematic diagram of the following gait principle of the quadrupedal intelligent following robot of the present invention.

[0028] Figure 5 This is a schematic diagram illustrating the following gait principle of the quadrupedal intelligent following robot of the present invention from the beginning to the second step.

[0029] Figure 6 This is a schematic diagram illustrating the following gait principle of the quadruped intelligent following robot of the present invention from the third to the fifth step.

[0030] Figure 7 This is a schematic diagram illustrating the following gait principle of the quadrupedal intelligent following robot of the present invention from step six to step eight.

[0031] Figure 8 This is a schematic diagram of the following gait principle of the quadruped intelligent following robot of the present invention from step nine to step eleven.

[0032] Figure 9 This is a schematic diagram of the twelfth step of the following gait principle of the quadruped intelligent following robot of the present invention.

[0033] Figure 10 This is a schematic diagram illustrating the following gait principle of the hexapod intelligent following robot of the present invention.

[0034] Figure 11 This is a schematic diagram illustrating the step-by-step following gait principle of the quadruped intelligent following robot of the present invention. Detailed Implementation

[0035] The following discussion is a recommended embodiment of the present invention, but is not limited to the following embodiments.

[0036] In practical applications, please refer to Figure 1 As shown, the present invention relates to an all-terrain environment-adaptive intelligent following robot, which includes a person being followed 100, a gait detection mechanism for the person being followed 200, and an intelligent following robot 300.

[0037] In a preferred embodiment of the present invention, the gait detection mechanism 200 measures the position and pressure information of the forefoot and heel of the followed person's feet in real time and sends this information to the intelligent following robot body 300 in real time.

[0038] In practical applications, please refer to Figure 2. The gait detection mechanism 200 of the person being followed involved in this invention includes a forefoot positioning and attitude measurement unit 211, a forefoot pressure sensor group 212, a heel pressure sensor group 213, a heel positioning and attitude measurement unit 214, a communication module 215, a clear and easily identifiable image following mark 216, a laser following mark 217, and a laser receiving sensor 218.

[0039] In a recommended embodiment of the present invention, please refer to Figure 2 As shown, in the gait detection mechanism 200, 211-215 are installed on the foot of the person being followed 100 in the form of insoles; 216-218 are installed on the heel of the person being followed 100's shoe.

[0040] In a preferred embodiment of the present invention, the positioning and attitude measurement units 211 and 214 include an inertial measurement unit, which can achieve precise millimeter-level positioning over short distances.

[0041] In a preferred embodiment of the present invention, pressure sensors 211 and 213 integrate microelectronics and micromachining technologies, combining micromechanical structures with electronic circuits to achieve sensitive monitoring and precise conversion of pressure changes.

[0042] In a preferred embodiment of the present invention, the communication module 215 implements wireless communication based on WIFI or Bluetooth.

[0043] In a preferred embodiment of the invention, the image following mark 216 and the laser following mark 217 are obvious, easily identifiable images that can be mounted on the heel of the person being followed, such as a 5cm x 5cm red capital letter, such as "L" or "R".

[0044] In a preferred embodiment of the invention, the image following marker 216 and the laser following marker 217 are prominent, easily identifiable LED luminescent images that can be mounted on the heels of the person being followed for identification at night.

[0045] The intelligent following robot body 300 receives real-time measurements of the position and pressure information of the forefoot and heel of the followed person's feet from the gait detection mechanism 200. Using this information, it determines the position, posture, and force control of its own feet, thus achieving a fully following mode and avoiding route planning and obstacle avoidance in complex road conditions.

[0046] In a preferred embodiment of the invention, such as Figure 3As shown, the intelligent following robot 300 completes walking planning through the walking planning device 330, realizes walking in a completely foot-following manner, completes walking through the walking execution device 340, adjusts its own center of gravity through the luggage rack controller 320, and realizes audio command reception through the auxiliary component 350. All the above devices are completed under the control of the main controller 310 and its included sub-modules.

[0047] In a preferred embodiment of the present invention, the intelligent following robot 300 adopts a quadrupedal walking mode such as a quadrupedal robot dog, a robot donkey, or a robot mule.

[0048] In a preferred embodiment of the present invention, the main controller 310 of the intelligent following robot 300 includes a communication module 311, which receives gait position and posture information sent from the gait detection mechanism 200 worn by the person being followed; a walking control module 312 controls a walking planning device 330 and a walking execution device 340 to complete foot position planning and force control, thereby achieving gait following of the person being followed; and a center of gravity controller 313 controls a luggage rack controller 320 to adjust the center of gravity during walking, thereby achieving stable walking.

[0049] In a recommended embodiment of the present invention, in addition to receiving gait position and posture information sent by the gait detection mechanism 200 worn by the person being followed, the communication module 311 also sends commands such as "pause" and "follow up" issued by the person being followed, or receives a distress signal issued by the intelligent following robot when the video analysis detects an obstacle.

[0050] In a preferred embodiment of the present invention, the communication module 311 implements wireless communication based on WIFI or Bluetooth.

[0051] In a preferred embodiment of the present invention, the walking planning module of the intelligent following robot 300 includes a high-definition camera 331, a video analysis module 332, a positioning and attitude measurement unit 333, a laser sensor 334, a binocular vision sensor 335, a GNSS / RTK 336, a UWB 337, a foot pressure sensor 338, and a radar ranging module 339.

[0052] In a recommended embodiment of the present invention, the communication module 311 receives the geographical location of the feet of the person being followed from the gait detection mechanism, including the horizontal displacement and height displacement relative to the starting position, the angle between the foot and the horizontal plane as the desired foot position, the difference between the position of its own foot and the desired foot position measured by the positioning and posture measuring unit 333 as the desired displacement, the desired vertical acceleration is calculated based on the vertical height difference between the current foot position and the adjacent forward foot position of the person being followed, the desired horizontal acceleration is calculated based on the state of walking, starting or stopping action, and the force control applied to each joint position is determined by combining its own weight, the force distribution received on the feet of the person being followed, and other information.

[0053] In a recommended embodiment of the present invention, the video analysis module 332 analyzes the video stream captured by the high-definition camera 331 in real time to locate the position of the laser following mark 217, and the laser emitter 334 emits a laser signal in real time to the laser receiving sensor 218 at the location of the laser following mark 217 on the back foot of the person being followed for distance measurement and positioning, thereby eliminating the measurement error accumulated by the positioning and attitude measurement unit 333 in real time and improving the following accuracy.

[0054] In a recommended embodiment of the present invention, the laser emitter 334 integrates a gimbal device, and the video analysis module 332 analyzes the video stream captured by the high-definition camera 331 in real time. After locating the position of the laser follower mark 217, the emission angle of the laser emitter 334 is adjusted, and the laser signal is emitted after being aligned with the laser receiving sensor adjacent to the laser follower mark 217.

[0055] In a recommended embodiment of the present invention, for an image frame containing the laser follower marker 217, the intelligent following robot 300 has recorded its pixel coordinates. Using back projection, the three-dimensional laser point cloud corresponding to the pixel coordinates can be obtained from the three-dimensional laser point cloud frame data at the corresponding moment of the image frame. These point cloud coordinates are the relative three-dimensional position coordinates of the laser follower marker 217.

[0056] In a preferred embodiment of the invention, both the laser emitter 334 and the laser receiver sensor 218 are mounted on the side of the intelligent following robot, and the laser receiver sensor 218 mounted on the gait detection mechanism 200 of the followed person's feet becomes a reflector. In this embodiment, the emission, reception, ranging, and positioning of the laser signal are all completed on the side of the intelligent following robot. The reflector of the gait detection mechanism 200 mounted on the followed person's feet is only used to reflect the laser signal. The results of laser ranging and positioning eliminate the accumulated measurement errors of the positioning and attitude measurement unit 333 and the positioning and attitude measurement units 211 and 214 of the gait detection mechanism in real time, thereby improving the following accuracy.

[0057] In a preferred embodiment of the present invention, the transmission, reception, ranging, and positioning of the laser signal are all completed on the side of the intelligent following robot. The reflector of the gait detection mechanism 200 installed on the feet of the person being followed is only used to reflect the laser signal. The reflector is a stainless steel, silver, mirror-like material ornament to achieve a high laser reflectivity, thereby achieving high-precision positioning.

[0058] In a recommended embodiment of the present invention, the transmission, reception, ranging, and positioning of the laser signal are all completed on the side of the intelligent following robot. The reflector of the gait detection mechanism 200 installed on the feet of the person being followed is only used to reflect the laser signal. The reflector is a 3M™ reflective film, a microcrystalline prism array reflective sheet, or a total station-specific reflective sheet to achieve a high laser reflectivity, thereby achieving high-precision positioning.

[0059] In a recommended embodiment of the present invention, the binocular vision sensor 335 locates the image tracking mark 216 on the heel of the person being followed in real time, detects the distance and position of the intelligent following robot to the image tracking mark in real time, and corrects the accumulated measurement errors of the positioning and attitude measurement unit 333 of the intelligent following robot and the positioning and attitude measurement units 211 and 214 of the gait detection mechanism.

[0060] In a preferred embodiment of the present invention, the image tracking tag 216 on the hind foot of the person being followed uses AprilTag technology to achieve high-precision positioning.

[0061] In a recommended embodiment of the present invention, the image tracking standard 216 and laser tracking mark 217 on the hind foot of the person being followed can be installed as an ankle ring worn on the ankle, or moved up to the knee to form a knee ring similar to a knee brace in order to prevent grass from obscuring them, and the position difference between the wearing position and the foot of the person being followed is compensated by a correction algorithm.

[0062] In a recommended embodiment of the present invention, the intelligent following robot 300 employs a relatively complete following strategy as follows: the initial positions x, y, z of the positioning and attitude measuring unit 333 on the intelligent following robot 300 are 0, 0, 0; the initial positions x, y, z of the positioning and attitude measuring units 211 and 214 installed on the gait detection mechanism 200 are the x, y, z values ​​of the relative position of the foot end of the laser following mark correction device measured by the laser positioning system based on the laser emitter 334 and the laser receiver sensor 218; during the walking process of the followed person 100, the difference between the x, y, z values ​​of the positioning and attitude measuring units 211 and 214 installed on the gait detection mechanism 200 and the positioning and attitude measuring unit 333 installed on the intelligent following robot 300 is the relative position coordinate; the intelligent following robot 300 uses the trajectory formed by the relative position coordinates of the gait detection mechanism 200 as the desired foot trajectory and plans its own walking execution device 340 to follow; when the followed person 100 is located at the emission / receiving position of the laser emitter 334 or the binocular vision sensor 335... When within the sensing range, the intelligent following robot 300 continuously senses the laser / image following marker affixed to the heel of the person being followed 100 and determines its three-dimensional position relative to the robot. It then continuously uses this information to correct the x, y, z values ​​of the positioning and attitude measuring units 211 and 214 on the gait detection mechanism 200, eliminating the accumulated relative measurement errors of the positioning and attitude measuring units 211 and 214 and the positioning and attitude measuring unit 333. When the person being followed 100 moves out of the emission / sensing range of the laser emitter 334 or the binocular vision sensor 335, the intelligent following robot 300 assumes that there are currently no accumulated measurement errors in the positioning and attitude measuring units 211 and 214 in the gait detection mechanism 200 and the positioning and attitude measuring unit 333 on the intelligent following robot 300, and continues to use the x, y, z values ​​of both. The difference of z is the expected displacement. The expected vertical acceleration is calculated based on the vertical height difference between the current foot position of the followed person and the adjacent foot position in front. The expected horizontal acceleration is calculated based on the state of movement, start or stop. Combined with its own weight and the force distribution on the feet of the followed person, the force control applied to each joint position is determined until the followed person 100 returns to the emission / sensing range of the laser emitter 334 or the binocular vision sensor 335.

[0063] In a recommended embodiment of the present invention, the person being followed 100 restricts their range of movement relative to the intelligent following robot 300 to ensure that the time of loss of contact within the sensing range does not exceed a short period of time, such as 15 seconds.

[0064] In a preferred embodiment of the present invention, the image following mark 216 and the laser following mark 217 may be combined into a single image mark.

[0065] In another recommended embodiment of the present invention, the gait detection mechanism 200 and the intelligent following robot 300 integrate GNSS / RTK 336 and obtain relative positions based on the difference of their respective world coordinates, thereby eliminating the measurement errors accumulated by the positioning and attitude measurement unit in real time and improving the following accuracy.

[0066] In another recommended embodiment of the invention, the gait detection mechanism 200 and the intelligent following robot 300 integrate UWB 337 to eliminate the measurement error accumulated by the positioning and attitude measurement unit in real time, thereby improving the following accuracy.

[0067] In another recommended embodiment of the present invention, the gait detection mechanism 200 and the intelligent following robot 300 integrate UWB 337 to eliminate the measurement errors accumulated by the positioning and attitude measurement unit in real time, thereby improving the following accuracy. Beacons are deployed near the environment where the followed person 100, the followed person gait detection mechanism 200, and the intelligent following robot 300 are located, so as to achieve the accuracy of eliminating the measurement errors accumulated by the positioning and attitude measurement unit in real time through UWB 337.

[0068] In a preferred embodiment of the present invention, the intelligent following robot 300 receives in real time the gait position information and posture information of the followed person 100 sent by the gait detection mechanism 200 through the communication module 331, as the foot position and posture desired by the intelligent following robot itself.

[0069] In a preferred embodiment of the present invention, the intelligent following robot 300 analyzes the video stream captured by the high-definition camera 331 in real time through the video analysis module 332 and, in conjunction with the radar ranging module 339, detects possible obstacles that may be inserted into the travel path in real time.

[0070] In a preferred embodiment of the present invention, the high-definition camera 331 has infrared night vision function, which can still capture clear video images when following a walk at night, and, in conjunction with the radar ranging module 337, find possible obstacles that may be inserted into the walking path.

[0071] In a preferred embodiment of the invention, the high-definition camera 331 has a gimbal equipped with a motor and a control system to achieve stable video quality.

[0072] In a recommended embodiment of the present invention, the walking planning device 330 of the intelligent following robot 300 adopts pure laser positioning based on laser emitter 334, laser following marker 217 and laser receiver 218, or pure visual positioning based on binocular vision sensor 335. This eliminates the inertial navigation system based on positioning and attitude measurement units 211 and 214 in the gait detection mechanism and foot positioning and attitude measurement unit 333 in the intelligent following robot 300. While simplifying the system complexity, blind spots may occur due to corners or steps obstructing the laser / image following marker. This is suitable for relatively simple open environment scenarios.

[0073] In a preferred embodiment of the present invention, the walking execution module 340 of the intelligent following robot 300 includes a rigid link 341 of the robot leg, a status register 342, an execution motor 343, a hip joint 344, a hip joint 345, a knee joint 346, and an ankle joint 347. The walking control module 312 reads the position difference between its own foot position information measured by the positioning and attitude measurement unit 333 and the foot position information of the person being followed measured by the expected gait detection mechanism stored in the status register 342, and calculates the torque of all joints, including the hip joint 346, hip joint 347, knee joint 348, and ankle joint 349, and controls the output of the execution motor 343. Through the superposition of the forces of the rigid link 341, complete foot-following walking is achieved in various situations.

[0074] In one preferred embodiment of the invention, the foot of the intelligent following robot 300 adopts the shape of the foot of a person wearing hiking shoes to completely simulate the shape of the foot of the person being followed.

[0075] In a preferred embodiment of the present invention, the bottom of the foot of the intelligent following robot 300 and the sole of the shoe of the person being followed have the same or substantially the same deformation force and friction force.

[0076] In a preferred embodiment of the present invention, the positioning and attitude measurement unit 333 of the intelligent following robot 300 includes a forefoot positioning and attitude measurement unit and a heel positioning and attitude measurement unit; the foot pressure sensor 338 includes a forefoot pressure sensor group and a heel pressure sensor group. These components are mounted on the foot of the intelligent following robot 300 in the form of components in an insole.

[0077] In a recommended embodiment of the present invention, the forefoot positioning and attitude measuring unit, the heel positioning and attitude measuring unit, the forefoot pressure sensor group, the heel pressure sensor group of the intelligent following robot 300 and the positioning and attitude measuring units 211 and 214 and pressure sensors in the following person gait detection mechanism 200 worn by the following person 100 are arranged in the same position.

[0078] In a preferred embodiment of the present invention, the positioning and attitude measurement units 211 and 214 in the gait detection mechanism 200 can be merged into a single positioning and attitude measurement unit to simplify system complexity.

[0079] In a recommended embodiment of the present invention, the walking control module 313 in the intelligent following robot 300 measures the foot pressure sensor 338 in real time. If the pressure is inconsistent with the expected pressure calculated based on its own weight, horizontal acceleration, vertical acceleration, and force control of each joint, that is, if the foot sinks due to the soft road surface, the walking control module 313 will compensate in time to achieve stable forward movement.

[0080] In a preferred embodiment of the present invention, the auxiliary device 350 of the intelligent following robot 300 includes a microphone 351 and an audio analysis module 352. The intelligent following robot 300 analyzes the audio information collected by the microphone 351 in real time through the audio analysis module 352, compares it with the pre-recorded voiceprint information of the person being followed 100, and analyzes and searches for possible commands such as "start" and "pause" from the person being followed in the matching audio information.

[0081] In a recommended embodiment of the present invention, the person being followed 100 operates a smart device APP such as a smartphone, smartwatch, or bracelet to send commands such as "start" or "pause" to the intelligent following robot 300 via wireless communication methods such as WIFI or Bluetooth.

[0082] In a preferred embodiment of the present invention, the luggage controller module of the intelligent following robot 300 includes a luggage rack rail 321, a luggage tray 322, an actuator motor 323, and a status register 324.

[0083] In a recommended embodiment of the present invention, the gait posture adopted by the intelligent following robot 300 during movement ensures that at any time, at least three points of the foot form a support surface. The center of gravity control module 312 calculates the desired position of the luggage tray 322, reads the current position stored in the status register 324, calculates the position difference, adjusts the luggage rack stepper motor 323, and adjusts the position of the luggage tray 322 along the luggage rack guide rail 321, so that the center of gravity is located within the geometric area of ​​the support surface, thereby achieving stable walking.

[0084] In a recommended embodiment of the present invention, when the intelligent following robot 300 takes each step forward, at least three of its feet remain stationary, forming a support plane with the stationary foot support points. The center of gravity of the intelligent following robot is controlled by the center of gravity control module 312 to keep it within the support plane, thereby achieving stability during walking. In a recommended embodiment of the present invention, the center of gravity control module 312 of the intelligent following robot 300 controls the robot with different strategies under different horizontal and vertical accelerations. For example, when accelerating forward, it controls the actuator motor 323 to move the luggage tray 322 forward along the luggage rack guide rail 321, so that the center of gravity tilts forward. Or, when going uphill, it controls the actuator motor 323 to move the luggage tray 322 to the left or right along the luggage rack guide rail 321, so that the center of gravity tilts to the side where the force is applied.

[0085] In a recommended embodiment of the present invention, a typical schematic diagram of the intelligent following robot 300 for starting, walking, and stopping is shown below. Figures 4 to 9 As shown, Before activation, both the person being followed and the intelligent following robot are located in a flat area. Within this area, the person being followed, wearing the gait detection device 200, stands at position "1. Start," with their left and right feet positioned at "Left 1" and "Right 1," respectively. The person wearing the gait detection device 200 then begins to walk naturally, taking the first step, the second step, and so on until the seventh step, after which they stop. Figure 4 sequence shown.

[0086] In a preferred embodiment of the present invention, the intelligent following robot, assuming the ground is perfectly flat, autonomously determines its footprint navigation within a flat area, such as... Figure 4 As shown. Arrived via autonomous navigation. Figure 5 The positions shown as "1. Start" are: left front at left 2, left back at left 1, right front at right 2, and right back at right 1.

[0087] In a recommended embodiment of the invention, the followed person's "3. Second step: Right 2" is not a normal step number, but rather falls at the same level as the left foot, such as... Figure 4 As shown in the "right 2" position, this allows the intelligent following robot 300 to have a more standard starting gait in a quadrupedal mode in a recommended embodiment of the present invention.

[0088] like Figure 5 As shown, in a recommended embodiment of the present invention, after the intelligent following robot enters the "1. Start" position, the foot position of the person being followed in the dotted area in front is the desired foot position in front.

[0089] like Figure 5As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its left front foot, moving from left 2 to left 3, and enters the state "2. First step", that is: left front is located at left 3, left rear is located at left 1, right front is located at right 2, and right rear is located at right 1. When the intelligent follower robot 300 steps forward with its left front foot, the three supporting feet are: left 1, right 1, and right 2.

[0090] like Figure 5 As shown, in a recommended embodiment of the present invention, the intelligent following robot 300 steps forward with its left rear foot, moving from left 1 to left 2, and enters the state "3. Second step", that is: left front is located at left 3, left rear is located at left 2, right front is located at right 2, and right rear is located at right 1. When the intelligent following robot 300 steps forward with its left rear foot, the three supporting feet are: left 3, right 1, and right 2.

[0091] like Figure 6 As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its right front foot, moves from right 2 to right 3, and enters the state "4. Third step", that is: left front is located at left 3, left rear is located at left 2, right front is located at right 3, and right rear is located at right 1. When the intelligent follower robot 300 steps forward with its right front foot, the three supporting feet are: left 3, left 2, and right 1.

[0092] like Figure 6 As shown, in a recommended embodiment of the present invention, the intelligent following robot 300 steps forward with its right rear foot, moving from right 1 to right 2, and enters the state "5. Fourth step", that is: left front is located at left 3, left rear is located at left 2, right front is located at right 3, and right rear is located at right 2. When the intelligent following robot 300 steps forward with its right rear foot, the three supporting feet are: left 3, left 2, and right 3.

[0093] like Figure 6 As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its left front foot, moving from left 3 to left 4, and enters the state "6. Fifth step", that is: left front is located at left 4, left rear is located at left 2, right front is located at right 3, and right rear is located at right 2. When the intelligent follower robot 300 steps forward with its left front foot, the three supporting feet are: left 2, right 2, and right 3.

[0094] like Figure 7 As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its left rear foot, moving from left 2 to left 3, and enters the state "7. Sixth step", that is: left front is located at left 4, left rear is located at left 3, right front is located at right 3, and right rear is located at right 2. When the intelligent follower robot 300 steps forward with its left rear foot, the three supporting feet are: left 4, right 2, and right 3.

[0095] like Figure 7As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its right front foot, moving from right 3 to right 4, and enters the state "8. Seventh step", that is: left front is located at left 4, left rear is located at left 3, right front is located at right 4, and right rear is located at right 2. When the intelligent follower robot 300 steps forward with its right front foot, the three supporting feet are: left 4, left 3, and right 2.

[0096] like Figure 7 As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its right rear foot, moving from right 2 to right 3, and enters the state "9. Eighth step", that is: left front is located at left 4, left rear is located at left 3, right front is located at right 4, and right rear is located at right 3. When the intelligent follower robot 300 steps forward with its right rear foot, the three supporting feet are: left 4, left 3, and right 4.

[0097] like Figure 8 As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its left front foot, moving from left 4 to left 5, and enters the state "10. Ninth Step", that is: left front is located at left 5, left rear is located at left 3, right front is located at right 4, and right rear is located at right 3. When the intelligent follower robot 300 steps forward with its left front foot, the three supporting feet are: left 3, right 3, and right 4.

[0098] like Figure 8 As shown, in a recommended embodiment of the present invention, the intelligent following robot 300 steps forward with its left rear foot, moving from left 3 to left 4, and enters the state "11. Tenth step", that is: left front is located at left 5, left rear is located at left 4, right front is located at right 4, and right rear is located at right 3. When the intelligent following robot 300 steps forward with its left rear foot, the three supporting feet are: left 5, right 3, and right 4.

[0099] like Figure 8 As shown, in a recommended embodiment of the present invention, the intelligent follower robot 300 steps forward with its right front foot, moving from right 4 to right 5, and enters the state "12. Eleventh step", that is: left front is located at left 5, left rear is located at left 4, right front is located at right 5, and right rear is located at right 3. When the intelligent follower robot 300 steps forward with its right front foot, the three supporting feet are: left 5, left 4, and right 3.

[0100] like Figure 9 As shown, in a recommended embodiment of the present invention, the intelligent following robot 300 steps forward with its right rear foot, moving from right 3 to right 4, and enters the state "13. Twelfth step", that is, the final step: left front is located at left 5, left rear is located at left 4, right front is located at right 5, and right rear is located at right 4. When the intelligent following robot 300 steps forward with its right rear foot, the three supporting feet are: left 5, left 4, and right 5.

[0101] In a recommended embodiment of the present invention, when the intelligent following robot 300 steps forward with one foot and the other three feet form supporting feet, the walking control module 313 uses the zero torque point (ZMP) algorithm to determine and maintain the dynamic stability of the intelligent following robot in motion, thereby achieving stable walking.

[0102] In another recommended embodiment of the present invention, when the intelligent following robot 300 steps forward with one foot and the other three feet form supporting feet, the walking control module 313 uses a model predictive control (MPC) algorithm to determine and maintain the dynamic stability of the intelligent following robot in motion, thereby achieving stable walking.

[0103] In a recommended embodiment of the present invention, the walking control module 313 calculates the desired horizontal acceleration for each stage, including starting from the initial state and stopping from the walking state, based on the desired horizontal movement speed. In this way, it adjusts the force control of a total of 12 joints, including the hip joint, hip joint, knee joint, and ankle joint of the three supporting feet, to achieve force control from a stationary state to a horizontal acceleration state for walking, thereby enabling stable walking.

[0104] In a recommended embodiment of the present invention, the walking control module 313 calculates the vertical acceleration during walking based on the height of the desired foot position after the next step and the height of the current position of the three supporting feet. In this way, it adjusts the force control of a total of 12 joints, including the hip joint, hip joint, knee joint, and ankle joint of the three supporting legs, to achieve force control in a height-elevated state and to walk stably.

[0105] In a recommended embodiment of the present invention, when the load is light, the walking control module 313 adopts larger strides during fast walking, such as when alternating between left and right, the front foot takes a step to the right of the next sequence's footprint, such as "1. Starting position, i.e., left front is left 2, left rear is left 1, right front is right 2, right rear is right 1" → "2. First step, i.e., left front is left 3, left rear is left 1, right front is right 2, right rear is right 1" → "3. Second step, i.e., left front is left 3, left rear is left 2, right front is right 2, right rear is right 1" → "4. Third step, i.e., left front is left 3, left rear is left 2, right front is right 4, right rear is right 1" → "5. Fourth step, i.e., left front is left 3, left rear is left 2, right front is right 4, right rear is right 3" → "6. The fifth step, namely, left front is left 5, left rear is left 2, right front is right 4, right rear is right 3, and so on, is similar. The essence of fast walking is another walking strategy based on completely following the footprints left by the person being followed.

[0106] In a recommended embodiment of the present invention, the walking control module 313 may adopt another stepping mode during walking, such as left front → right front → left back → right back, but in essence, it completely follows the footprints left by the person being followed 100.

[0107] In a recommended embodiment of the present invention, when the person being followed 100 is walking on a steep hillside, if he or she takes a step over multiple steps or adopts a "straight-line gait" (i.e., the person's feet land on the same sagittal axis and the stride width approaches zero), and the intelligent following robot 300's support surface narrows and it becomes difficult to stably control its center of gravity to complete the following, the intelligent following robot 300 will notify the person being followed by means of a buzzer or the like.

[0108] In another recommended embodiment of the present invention, when the intelligent following robot 300 is unable to complete the following, it notifies the person being followed via an APP notification on an electronic device such as a smartphone, smartwatch, or bracelet held by the person being followed.

[0109] In a recommended embodiment of the present invention, after receiving a notification that the intelligent following robot 300 is having difficulty passing through, the person being followed 100 can use a smartphone, watch, or other device APP to select to return to a pair of feet that were previously in a stable position, and then return to this stable position and accurately place the left and right feet on the designated foot positions before starting a new natural walking process.

[0110] In another recommended embodiment of the present invention, in application scenarios with flat road surfaces such as cities, factories, schools, and squares, the intelligent following robot 300 uses an electric motor to output torque, which is then transmitted to the drive shaft after being reduced and increased by gears. The drive shaft drives the tires to rotate, thus enabling the vehicle to move forward.

[0111] In another recommended embodiment of the present invention, on roads where humans need to use various climbing skills to pass through, such as steep mountains and valleys full of large rocks, the intelligent following robot 300 can move forward by using multiple flight modes such as fixed-wing, rotor, tilt rotor, and flapping wing.

[0112] In another recommended embodiment of the present invention, the intelligent following robot 300 adopts a composite mode including quadruped, hexapod, multi-legged walking, wheel drive, and flight to achieve vehicle forward movement, and can switch between different road conditions to achieve a balance in terms of speed, safety, passability, energy consumption, etc.

[0113] In a recommended embodiment of the present invention, the person being followed 100 may pass through narrow steps that can only accommodate the front half of a foot while walking, where the pressure difference between the forefoot and the heel is large. Based on this information, the intelligent following robot 300 adjusts the force control of all joints, including the ankle joint, to achieve stable walking.

[0114] In a recommended embodiment of the present invention, the person being followed 100, after systematic training, understands that the intelligent following robot 300 is unable to climb with the same precision and agility as a human. When encountering too narrow a foothold or too steep a road, the person being followed 100 chooses other walking routes with more stable, solid, and spacious footholds, or chooses to switch to other driving modes, such as the flight mode mentioned above, for a short period of time.

[0115] In another recommended embodiment of the present invention, when the gait detection mechanism 200 and the intelligent following robot 300 obtain world coordinates by fusing GNSS / RTK / UWB, the intelligent following robot 300 receives all foot information of the followed person 100 from the starting point to the destination at once. Without the need for the followed person 100 and the followed person gait detection mechanism 200 module, it can directly achieve weight-bearing forward movement in composite modes including quadrupedal, hexapod, multi-legged walking, wheeled drive, and flight. Combined with the high-definition camera 342, video analysis module 344, and radar ranging module 348, it can detect possible obstacles that may be inserted into the travel path in real time, and can reach the destination autonomously.

[0116] In another recommended embodiment of the invention, such as Figure 10 As shown, the walking actuator of the intelligent following robot 300 is in the form of a six-legged shape.

[0117] Before activation, both the person being followed and the intelligent following robot are located in a flat area. Within this area, the person being followed, wearing the intelligent following robot 1 with a gait detection mechanism, stands at position "1. Person Being Followed 100 Start," with their left and right feet positioned at "Left 1" and "Right 1," respectively. The person being followed 100, wearing the gait detection mechanism 200, begins to walk naturally, taking the first step, the second step, up to the seventh step, and then stopping. Figure 10 sequence shown.

[0118] like Figure 10 As shown, in another recommended embodiment of the invention, the intelligent following robot, assuming the ground is perfectly flat, autonomously determines its footprint navigation within a flat area until it reaches its destination. Figure 10 The positions shown for "1. Intelligent Follower Robot 300 Start" are: left front is 3rd from the left, left center is 2nd from the left, left rear is 1st from the left, right front is 3rd from the right, right center is 2nd from the right, and right rear is 1st from the right.

[0119] like Figure 10 As shown, in another recommended embodiment of the present invention, the steps “3. Second step: Right 2” and “5. Fourth step: Right 3” of the followed person are not normal steps, but fall at the same level as the left foot, as shown in the “Right 2” and “Right 3” positions in the figure, respectively, so that the intelligent following robot has a more standard starting gait in the hexapod mode in another recommended embodiment of the present invention.

[0120] like Figure 10 As shown, in another recommended embodiment of the present invention, after the intelligent following robot enters the "1. Intelligent following robot 300 start" position, the foot position of the person being followed in the dotted area in front is the desired foot position in front.

[0121] In the recommended embodiment of the hexapod walking of the present invention, the subsequent action steps of the intelligent following robot 300 are similar to those of the quadruped robot described above, as follows. The intelligent follower robot 300 continues moving forward, stepping forward with its left front foot, moving from left 3 to left 4. During this step, the five supporting feet are: left 2, left 1, right 3, right 2, and right 1. The intelligent follower robot 300 continues moving forward, stepping forward with its left middle foot, moving from left 2 to left 3. During this step, the five supporting feet are: left 4, left 1, right 3, right 2, and right 1. The intelligent follower robot 300 continues moving forward, stepping forward with its left rear foot, moving from left 1 to left 2. During this step, the five supporting feet are: left 4, left 3, right 3, right 2, and right 1. The intelligent follower robot 300 continues moving forward, stepping forward with its right front foot, moving from right 3 to right 4. During this step, the five supporting feet are: left 4, left 3, left 2, right 2, and right 1. The intelligent follower robot 300 continues moving forward, stepping forward with its right middle foot, moving from right 2 to right 3. During this step, the five supporting feet are: left 4, left 3, left 2, right 4, and right 1. The intelligent follower robot 300 continues moving forward, stepping forward with its right rear foot, moving from right 1 to right 2. During this step, the five supporting feet are: left 4, left 3, left 2, right 4, and right 3. The intelligent follower robot 300 continues moving forward, stepping forward with its left front foot, moving from left 4 to left 5. During the step, the five supporting feet are: left 3, left 2, right 4, right 3, and right 2. The intelligent follower robot 300 continues moving forward, stepping forward with its left middle foot, moving from left 3 to left 4. During the step, the five supporting feet are: left 5, left 2, right 4, right 3, and right 2. The intelligent follower robot 300 continues moving forward, stepping forward with its left rear foot, moving from left 2 to left 3. During this step, the five supporting feet are: left 5, left 4, right 4, right 3, and right 2. The intelligent follower robot 300 continues moving forward, stepping forward with its right front foot, moving from right 4 to right 5. During the step, the five supporting feet are: left 5, left 4, left 3, right 3, and right 2. The intelligent follower robot 300 continues moving forward, stepping forward with its right middle foot, moving from right 3 to right 4. During the step, the five supporting feet are: left 5, left 4, left 3, right 5, and right 2. The intelligent follower robot 300 continues to move forward, stepping forward with its right rear foot, moving from right 2 to right 3, and then stopping. During the stepping process, the five supporting feet are: left 5, left 4, left 3, right 5, and right 4.

[0122] In another recommended embodiment of the present invention, the walking actuator of the intelligent following robot 300 is in the form of a six-legged form, in which only one leg steps out at any time, while the remaining five legs form a support, providing a larger support surface and better stability.

[0123] In another recommended embodiment of the present invention, the walking actuator of the intelligent following robot 300 is in the form of a six-legged animal. The intelligent following robot 300 calculates the desired horizontal acceleration from the starting state based on the desired horizontal movement speed. At the same time, it calculates the desired vertical acceleration based on the height difference of the foot position in the sequence sent by the gait detection mechanism 200 of the followed person. It adjusts the force control of 20 joints, including the hip joint, hip joint, knee joint, and ankle joint of the five supporting legs, to achieve force control in the horizontal acceleration state from a stationary state to walking and in states such as height increase and decrease, thereby achieving stable walking.

[0124] In a recommended embodiment of the present invention, the walking actuator of the intelligent following robot 300 is in the form of a six-legged animal. When the intelligent following robot 300 steps forward with one leg and the remaining five legs form supporting legs, the walking control module 313 uses the zero torque point (ZMP) algorithm or the model predictive control (MPC) algorithm to judge and maintain the dynamic stability of the intelligent following robot in motion, so as to achieve stable walking.

[0125] In another recommended embodiment of the present invention, the walking actuator of the intelligent following robot 300 is quadrupedal and walks by taking steps in a staggered manner to better simulate the longer distance between the forelegs and hindlegs of quadrupedal animals.

[0126] Before activation, both the person being followed and the intelligent following robot are located in a flat area. Within this area, the person being followed, wearing the intelligent following robot 1 with a gait detection mechanism, stands at position "1. Person Being Followed 100 Start," with their left and right feet positioned at "Left 1" and "Right 1," respectively. The person being followed 100, wearing the gait detection mechanism 200, begins to walk naturally, taking the first step, the second step, up to the seventh step, and then stopping. Figure 11 sequence shown.

[0127] like Figure 11 As shown, in another recommended embodiment of the invention, the intelligent following robot, assuming the ground is perfectly flat, autonomously determines its footprint navigation within a flat area until it reaches its destination. Figure 11 The positions shown for "1. Intelligent Follower Robot 300 Start" are: left front at left 3, left rear at left 1, right front at right 3, and right rear at right 1.

[0128] like Figure 11 As shown, in another recommended embodiment of the present invention, the steps “3. Second step: Right 2” and “5. Fourth step: Right 3” of the followed person are not normal steps, but fall at the same level as the left foot, as shown in the “Right 2” and “Right 3” positions in the figure, respectively, so that the intelligent following robot has a more standard starting gait in the step-by-step quadruped mode in another recommended embodiment of the present invention.

[0129] like Figure 11 As shown, in another recommended embodiment of the present invention, after the intelligent following robot enters the "1. Intelligent following robot 300 start" position, the foot position of the followed person in the dotted area is the expected foot position in front.

[0130] In the recommended embodiment of the quadrupedal step-walking of the present invention, the subsequent action steps of the intelligent following robot 300 are similar to the action steps of the quadrupedal robot described above, as follows. The intelligent follower robot 300 continues moving forward, stepping forward with its left front foot, moving from left 3 to left 4. During the step, the three supporting feet are: left 1, right 1, and right 3. The intelligent follower robot 300 continues moving forward, stepping forward with its left rear foot, moving from left 1 to left 2. During the step, the three supporting feet are: left 4, right 1, and right 3. The intelligent follower robot 300 continues to move forward, stepping forward with its right front foot, moving from right 3 to right 4. During the step, the three supporting feet are: left 4, left 2, and right 1. The intelligent follower robot 300 continues moving forward, stepping forward with its right rear foot, moving from right 1 to right 2. During the step, the three supporting feet are: left 4, left 2, and right 4. The intelligent follower robot 300 continues moving forward, stepping forward with its left front foot, moving from left 4 to left 5. During the step, the three supporting feet are: left 2, right 2, and right 4. The intelligent follower robot 300 continues moving forward, stepping forward with its left rear foot, moving from left 2 to left 3. During the step, the three supporting feet are: left 5, right 2, and right 4. The intelligent follower robot 300 continues to move forward, stepping forward with its right front foot, moving from right 4 to right 5. During the step, the three supporting feet are: left 5, left 3, and right 2. The intelligent follower robot 300 continues to move forward, stepping forward with its right rear foot, moving from right 2 to right 3, and then stopping. During the stepping process, the three supporting feet are: left 5, left 3, and right 5.

[0131] In the recommended embodiment of the quadrupedal stepping walking of the present invention, the subsequent action steps of the intelligent following robot 300 adopt another biomimetic stepping method, stepping with the back foot first, as follows: The intelligent follower robot 300 continues moving forward, stepping forward with its left rear foot, moving from left 1 to left 2. During the step, the three supporting feet are: left 3, right 1, and right 3. The intelligent follower robot 300 continues moving forward, stepping forward with its right rear foot, moving from right 1 to right 2. During the step, the three supporting feet are: left 3, left 2, and right 3. The intelligent follower robot 300 continues moving forward, stepping forward with its left front foot, moving from left 3 to left 4. During the step, the three supporting feet are: left 2, right 3, and right 2. The intelligent follower robot 300 continues to move forward, stepping forward with its right front foot, moving from right 3 to right 4. During the step, the three supporting feet are: left 4, left 2, and right 2. The intelligent follower robot 300 continues to move forward, stepping forward with its left rear foot, moving from left 2 to left 3. During the step, the three supporting feet are: left 4, right 2, and right 4. The intelligent follower robot 300 continues to move forward, stepping forward with its right rear foot, moving from right 2 to right 3. During the step, the three supporting feet are: left 4, left 3, and right 4. The intelligent follower robot 300 continues moving forward, stepping forward with its left front foot, moving from left 4 to left 5. During the step, the three supporting feet are: left 3, right 4, and right 3. The intelligent follower robot 300 continues to move forward, stepping forward with its right front foot, moving from right 4 to right 5. During the step, the three supporting feet are: left 5, left 3, and right 3.

[0132] The above-mentioned gait patterns, including quadrupedal walking, hexapedal walking, and two types of quadrupedal alternating walking, are only four examples of technically feasible gait schemes. Based on these, many similar tracking gait schemes can be extended. In essence, the intelligent following robot 300 strictly keeps the foot position consistent with the followed person, thereby avoiding road surface modeling, route planning, and obstacle avoidance algorithms in complex road conditions.

[0133] In another recommended embodiment of the present invention, the foot of the intelligent following robot 300 does not have a foot like that of the person being followed 100, nor does it have an ankle joint. Its foot is a biomimetic form of hoofed creatures. During walking, the center of the hoof rests on the pressure center calculated from the area and pressure value of the forefoot and hindfoot of the person being followed. Since hoofed feet have a relatively small ground contact area, it is recommended to adopt a six-legged walking mode to obtain a larger overall ground contact area from five supporting feet, thereby increasing the upper limit of the load of the intelligent following robot 300 when it has the same ground pressure as the person being followed 100.

[0134] In another recommended embodiment of the present invention, the person being followed 100 can be a real horse, dog, or other walking or running creature, and the intelligent following robot 300 strictly follows the position of its feet, thereby avoiding road surface modeling, route planning, and obstacle avoidance algorithms in complex road conditions.

[0135] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0136] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution device.

[0137] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0138] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0139] In the description of this specification, references to terms such as "an embodiment," "some embodiments," "example," "specific example," or "embodiment," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0140] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

[0141] This invention employs the technical solution of an all-terrain adaptive intelligent following robot. The desired foot position is determined by the foot position and posture of the followed person. Vertical acceleration is calculated from the height difference of the followed person's sequential foot positions, and horizontal acceleration is calculated from the starting, accelerating, decelerating, and stopping states. Combined with the robot's own weight, the torque of all joints, including the hip, hip, knee, and ankle joints, is derived in reverse. In this invention, the visual analysis technology using a high-definition camera as input and the radar ranging module are only used to detect newly inserted obstacles on the desired travel route. The basic technical principle of this invention, and the foot-guided navigation mode, avoids the need for road surface modeling, route planning, and obstacle avoidance algorithms in complex road conditions. Combined with center of gravity control, this enables the intelligent following robot to achieve stable following under load in all-terrain environments, traversing complex terrains that are difficult for traditional wheeled and multi-legged robots to cross, such as steps, railway tracks, and shallow water-covered riverbeds. This demonstrates significant technical advantages and practical value.

[0142] In this specification, the invention has been described with reference to specific embodiments thereof. However, it will be apparent that various modifications and variations can be made without departing from the spirit and scope of the invention. Therefore, the specification and drawings should be considered illustrative rather than restrictive.

Claims

1. An all-terrain environment-adaptive intelligent following robot, characterized in that, The following processing steps are included: (1) The person being followed wears a gait detection device and walks naturally, actively choosing a suitable path, avoiding obstacles, and stepping on a relatively flat, stable, safe, and foot-accommodating location for safe walking; (2) The gait detection mechanism of the followed person records the gait information of the followed person's two feet and the pressure information of the body weight in real time, and sends this information to the intelligent following robot in real time; (3) Set the foot position and posture of the intelligent following robot to be completely consistent with the foot position and posture of the person being followed; (4) The intelligent following robot moves forward in various stages based on its own weight, the pressure ratio of the front and back feet of the person being followed, the current foot position and the expected foot position of the person being followed in front, including starting, accelerating, decelerating, stopping, going up and down slopes, and going up and down stairs. (5) The intelligent following robot achieves balance control during movement, including going up and down slopes, going up and down steps, starting, accelerating, decelerating and stopping, through center of gravity control.

2. The all-terrain environment-adaptive intelligent following robot according to claim 1, characterized in that, The gait information includes foot position information and foot pressure information, wherein, The foot position information includes the three-dimensional position information of the forefoot and heel of the left and right feet along the x, y, and z axes. The foot pressure information includes information on the pressure and weight-bearing ratio of the sole and heel.

3. The all-terrain environment-adaptive intelligent following robot according to claim 1, characterized in that, Step (3) specifically includes the following steps: (3.1) The intelligent following robot measures the position of its own feet through a positioning and attitude measurement unit; (3.2) The intelligent following robot receives the position of the foot of the person being followed, measured by the positioning and attitude measurement unit in the gait detection mechanism; (3.3) Determine the expected foot displacement of the intelligent following robot based on the position difference between the position of the robot's own foot and the position of the foot of the person being followed. (3.4) The intelligent following robot uses a high-definition camera and video analysis module to locate the laser tracking mark on the heel of the person being followed in real time, and uses a laser transmitting / receiving sensor to detect the relative coordinates of the intelligent following robot to the laser tracking mark in real time. After correction, the position of the foot of the person being followed is obtained, thereby correcting the accumulated measurement errors of the positioning and attitude measurement unit of the intelligent following robot and the positioning and attitude measurement unit of the gait detection mechanism; or The intelligent following robot uses binocular positioning to detect the relative coordinates of the robot to the image tracking marker in real time. After correction, it obtains the position of the followed person's feet, thereby correcting the accumulated measurement errors of the positioning and attitude measurement units of the intelligent following robot and the gait detection mechanism; or The gait detection mechanism and the intelligent following robot integrate GNSS / RTK positioning to correct the accumulated measurement errors of the positioning and attitude measurement units of the intelligent following robot and the gait detection mechanism in real time; or The gait detection mechanism and the intelligent following robot integrate UWB positioning to correct the accumulated measurement errors of the positioning and attitude measurement units of the intelligent following robot and the gait detection mechanism in real time.

4. The all-terrain environment-adaptive intelligent following robot according to claim 3, characterized in that, Step (4) specifically includes the following steps: (4.1) The intelligent following robot is 3-5 steps behind the person being followed. By analyzing the height difference between its current foot position and the foot position of the person being followed in front, the vertical acceleration of each foot of the intelligent following robot is calculated. (4.2) The intelligent following robot calculates the expected horizontal acceleration of its main body based on its own start, acceleration, deceleration and stop decisions, combined with its current foot position and the foot position of the person being followed in front of it. (4.3) The intelligent following robot calculates the torque of all joints, including the hip joint, hip joint, knee joint and ankle joint, in reverse by combining the desired vertical acceleration, pressure ratio and horizontal acceleration of each foot end and its own weight. (4.4) The intelligent following robot, through intelligent analysis of the high-definition camera and video analysis module, combined with radar wave ranging, discovers newly added obstacles inserted in the walking path of the followed person, adjusts the torque of all joints, including the hip joint, hip joint, knee joint and ankle joint, actively stops following and notifies the followed person of the current status. (4.5) The intelligent following robot receives instructions from the person being followed through the communication module, or senses the voice instructions of the person being followed through voice recognition, adjusts the torque of all joints, including the hip joint, hip joint, knee joint and ankle joint, starts or stops following and notifies the person being followed of the current status.

5. The all-terrain environment-adaptive intelligent following robot according to claim 3, characterized in that, Step (5) specifically includes the following steps: (5.1) The intelligent following robot controls its center of gravity by adjusting the position of the luggage rack relative to the body of the intelligent following robot, including the front, back, left and right positions of its back. (5.2) When the intelligent following robot takes each step forward, at least three of its feet remain stationary, forming a support plane with the stationary foot support points. By controlling the center of gravity of the intelligent following robot, it is kept within the plane formed by the support points. (5.3) The center of gravity control device of the intelligent following robot uses different strategies to control the center of gravity under different horizontal and vertical accelerations, including controlling the center of gravity to tilt forward when accelerating forward, or controlling the center of gravity to tilt to the side of the force exerted when going uphill.

Images