Multi-mode intelligent baby carriage system based on SLAM navigation and control method

Abstract

The invention discloses a multi-mode intelligent baby carriage system based on SLAM (Simultaneous Localization and Mapping) navigation and a control method, which combine SLAM navigation and a multi-mode data processing technology and realize multiple functions of autonomous navigation, environment perception, user following, autonomous adjustment, remote monitoring and the like of a baby carriage. According to the baby carriage, a scissor-fork type structure is adopted, autonomous adjustment is achieved, an intelligent shaking and music playing mode is arranged, natural shaking is simulated, and meanwhile the dynamic forward leaning function is achieved; through the SLAM navigation and deep learning method, autonomous obstacle avoidance, intelligent following, multi-mode interaction and personalized service are realized, and the intelligent voice interaction module realizes voice control of the baby carriage and also supports remote monitoring and real-time feedback functions. According to the multi-mode intelligent baby carriage, the intelligent level and safety of the baby carriage are improved, and more convenient and comfortable child-rearing experience is provided for parents.

Description

Technical Field

[0001] The present invention relates to the technical field of infant appliances, and particularly relates to an intelligent baby stroller system integrating artificial intelligence, multi-sensor fusion and mechanical design, and a SLAM autonomous navigation control method. Background Art

[0002] With the rapid development of intelligent technology, intelligent technology has penetrated into all aspects of our lives. As an important tool for infants and young children to travel, baby strollers are also gradually developing towards the intelligent direction. Traditional baby strollers have a single function and lack the ability to actively adapt to complex environments, making it difficult to meet the needs of modern families for intelligent parenting. In the prior art, although some patents have proposed basic intelligent control (such as temperature and pressure), navigation and obstacle avoidance (such as terrain analysis systems), and music playback functions, there are still problems such as insufficient navigation accuracy, lag in dynamic response, and single interaction function. In order to provide a safer, more convenient, and more comfortable travel experience, a multi-modal intelligent baby stroller system based on SLAM (Simultaneous Localization and Mapping) navigation has emerged. This baby stroller system involves fields such as dynamic structure optimization, artificial intelligence technology, SLAM navigation technology, and deep learning. Among them, the structure optimization adopts a modular design and a lightweight structure to achieve multi-dimensional adjustment functions, and is equipped with intelligent rocking and music playback modes to simulate natural rocking, and also has a dynamic forward tilt function to improve the stability and comfort of the baby stroller; with the help of artificial intelligence technology, voice control of the baby stroller is realized, and through instruction sending, intelligent control of the baby stroller is achieved; using SLAM navigation technology, integrating lidar, visual SLAM, IMU multi-sensors and path planning algorithms, autonomous obstacle avoidance and intelligent following are realized; deep learning technology endows the baby stroller with multi-modal interaction and personalized services. Combining with a deep learning model, an appeasement mode is recommended according to the baby's behavior data. For example, when it is detected that the baby is crying, the intelligent rocking function is activated; in addition, in terms of intelligent control, remote monitoring and real-time feedback functions are also supported to meet the parents' real-time attention needs for the baby's state. The proposal of the multi-modal intelligent baby stroller not only improves the intelligent level and safety of the baby stroller, but also provides a more convenient and comfortable parenting experience for parents.

[0003] Therefore, inventing a multi-modal intelligent baby stroller system with autonomous navigation and obstacle avoidance, automatic adjustment, and automatic appeasement has great market potential. This system combines advanced SLAM navigation technology, multi-modal interaction methods, and intelligent control algorithms, bringing a revolutionary change to the travel of infants and young children. Summary of the Invention

[0004] The present invention mainly overcomes the deficiencies in the prior art and provides a multi-modal intelligent baby stroller system and control method based on SLAM navigation, which improves the intelligence level of the baby stroller through innovative technologies.

[0005] To achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] The multi-modal intelligent baby stroller system based on SLAM navigation comprises the following components: a mobile chassis, a lifting device, a rocking device, a tilting device, a shading device, an intelligent monitoring system, an autonomous navigation and intelligent voice interaction system.

[0007] The mobile chassis is connected to the lifting device of the baby stroller through four threaded posts;

[0008] The rocking device is supported by a bearing seat and is connected to the lifting device;

[0009] The tilting device is connected to the rocking device by the pressure of four springs;

[0010] The shading device is connected to the safety seat through a rotating mechanism;

[0011] Furthermore, the mobile chassis includes a braking device, a chassis housing, threaded posts, a bottom plate, a cover plate, and a control core component. The braking device is connected to the driving wheels and is a foot-operated brake; the inner wall of the lower end of the chassis housing has reserved holes that are connected to the mounting holes of the bottom plate by bolts; the control core component of the baby stroller is placed on the upper end of the bottom plate, and four threaded posts are distributed around the bottom plate for support; the upper ends of the threaded posts are connected to the cover plate.

[0012] Furthermore, the lifting device includes a lower load-bearing platform, a middle load-bearing platform, bearing seats, linear guide rail sliders, a scissor lift module, a ball screw nut module, and a drive module. The linear guide rail slider module includes linear guide rails and sliders. The linear guide rails are respectively fixed on the middle and lower load-bearing platforms, and the sliders are fixed on the linear guide rails. The bearing seats are respectively fixed on the sliders and the middle and lower load-bearing platforms. The scissor lift module includes a scissor support frame, a reinforcing rod, bearings, and optical axis 1. The bearings are installed in the hole positions at both ends of the support frame, and optical axis 1 is used to connect the bearing seats and the bearings on the support frame. The reinforcing rod is fixed on the opposite support frames. The ball screw nut module includes a ball screw, a screw nut, a coupling, a stepping motor 1, and a fixing member. The stepping motor 1 is fixed on the lower load-bearing platform through a stepping motor bracket. The coupling connects the stepping motor and the ball screw. The fixing member connects the screw nut and the optical axis 1 connected to the slider. The drive module is fixed at the end of the stepping motor 1.

[0013] Further, the shaking device includes a middle load-bearing platform, an upper load-bearing platform, a rotating mechanism, a compression spring, and spring limit blocks. The rotating mechanism includes a stepper motor, a driving wheel, a driven wheel, a 2gt synchronous belt, a two-dimensional cloud platform, and bearings. The adjustable support base includes a lower support base, an upper support base, and a flange. The lower support base is connected to the bearing, the upper support base is connected to the lower support base through a positioning pin, the flange is connected to the upper support base through bolts, and the upper load-bearing platform is connected to the flange through bolts. The spring limit blocks are evenly arranged on the upper load-bearing platform and the middle load-bearing platform through bolt connections, and the upper and lower ends of the compression spring are respectively fixed to the spring limit blocks.

[0014] Further, the forward tilting device includes an upper load-bearing platform, an electric push rod, connecting bracket 1, connecting bracket 2, a fixing pin, a bearing seat, a light shaft 3, and a safety seat. Connecting bracket 1 is fixed to the upper load-bearing platform through bolts, and connecting bracket 2 is fixed to the safety seat through bolts. The electric push rod is fixed to connecting bracket 1 and connecting bracket 2 through a fixing pin. The bearing seat is fixed to the upper load-bearing platform through bolts, and the light shaft 3 connects the bearing seat and the safety seat.

[0015] Further, the light-shielding device includes a safety seat, a light-shielding cover, a rotating motor, and a light sensor. The light-shielding cover is arc-shaped and connected to the safety seat through a rotating shaft. The rotating motor is connected to the rotating shaft through a flange. The light sensor is connected to the Arduino core board to achieve automatic sunshading.

[0016] Further, the intelligent monitoring system includes a kognqizil camera, an industrial control computer, and an ESP8266. The camera is fixed above the safety seat to collect image information. The industrial control computer compares and analyzes the image information, and the ESP8266 sends information to the mobile phone / PC terminal to detect whether the baby turns over or the head is blocked, and alarms and sends information in a timely manner when there is a situation.

[0017] Further, the control core components include an Arduino core board, a servo motor driver and encoder, a servo motor, a driving wheel, and a universal wheel. The Arduino core board is placed at the center position on the top of the bottom plate. The servo motor driver and encoder are integrated around the Arduino core board. The servo motor is fixed to the motor bracket through bolts. The motor bracket is connected to the mounting holes on the bottom plate and symmetrically distributed at both ends under the bottom plate. The servo motor is fixed to the driving wheel through a bushing.

[0018] Furthermore, the autonomous navigation and intelligent voice interaction system of the stroller includes a SLAM algorithm, a path planning algorithm, and each hardware module. The SLAM algorithm constructs a real-time 3D environment map using an adaptive feature extraction algorithm. The path planning algorithm performs dynamic path planning using an improved A* algorithm. The hardware module includes a depth camera, a lidar, an IMU, a voice sensor, a temperature and humidity sensor, and an air quality sensor. The depth camera is fixed at the front end of the chassis shell. The lidar is fixed at the edge of the cover. The lidar sensor is connected to the lidar and fixed on the cover. The voice sensor and the IMU are fixed on the bottom plate. The temperature and humidity sensor and the air quality sensor are fixed on the side of the bottom plate shell.

[0019] A SLAM navigation method for an intelligent stroller includes at least:

[0020] S1: First, the stroller obtains perception data of the surrounding environment through the sensors it carries (lidar, depth camera, IMU);

[0021] S2: Use an adaptive feature point extraction technique to dynamically extract edge points and plane points from the acquired images;

[0022] S3: Use the perception data for self-localization to determine the position and attitude of the stroller in the environment;

[0023] S4: According to the environmental perception data and the result of self-localization, convert the perception data into a representation of the environment, and then combine visual ORB feature points to construct a multi-scale 3D environment map, laying a foundation for the next navigation and path planning;

[0024] S5: Use the perception data to update the map information in a timely manner to ensure the real-time and accuracy of the 3D environment map.

[0025] S6: Introduce an improved A* algorithm for dynamic path planning, detect dynamic obstacles in real time to filter out interference points, and adjust the path in real time to ensure the safety of the stroller's travel.

[0026] Beneficial effects:

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. The present invention relates to the technical field of strollers, especially an intelligent stroller system and control method that combines SLAM navigation technology and multi-modal data processing technology. By integrating multiple sensors and intelligent algorithms, the intelligence level of the stroller is improved, and its stability and reliability are enhanced;

[0029] 2. The present invention enhances the user experience. The navigation system based on SLAM technology enables the stroller to achieve autonomous path planning and user following in complex environments, reducing the burden on parents; the intelligent voice interaction system supports various methods such as voice communication and voice control, making the operation more convenient and intuitive; the remote status monitoring function monitors the baby's status in real time and sends the monitoring data to parents in real time; the automatic lifting and dynamic shaking functions meet the needs of parents of different heights and provide shaking and soothing, improving the baby's comfort and the parents' convenience. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the overall structure of the stroller of the present invention;

[0031] Figure 2 It is a schematic diagram of the structure of the mobile chassis of the present invention;

[0032] Figure 3 It is a schematic diagram of the scissor lift device of the present invention;

[0033] Figure 4 It is a schematic diagram of the shaking device of the present invention;

[0034] Figure 5 It is a schematic diagram of the forward tilt and light shielding device of the present invention;

[0035] Figure 6 It is a flowchart of the SLAM navigation and chassis control of the present invention;

[0036] Figure 7 It is a flowchart of the intelligent voice interaction system of the present invention;

[0037] In the figure: 1, mobile chassis; 2, scissor lift device; 3, shaking device; 4, forward tilt device; 5, light shielding device; 6, threaded column; 7, scissor lift module; 8, compression spring; 9, rotating mechanism; 10, safety seat; 11, braking device; 13, bottom plate; 14, cover plate; 15, driving wheel; 16, depth camera; 17, temperature and humidity sensor; 18, air quality sensor; 19, servo motor; 20, caster; 21, motor bracket; 22, camera; 23, lidar; 24, voice sensor; 25, lower load-bearing platform; 26, middle load-bearing platform; 27, bearing seat; 28, linear guide rail; 29, linear guide rail slider; 30, ball screw nut module; 31, stepper motor 1; 32, upper load-bearing platform; 33, spring limit block; 34, stepper motor 2; 35, 2gt synchronous belt; 36, electric push rod; 37, bearing seat 2; 38, optical axis 3; 39, light shielding cover; 40, rotating motor; 41, light sensor DETAILED DESCRIPTION OF THE INVENTION

[0038] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only a part of the embodiments of the present invention, rather than all embodiments.

[0039] Embodiment 1:

[0040] Referring to Figures 1-5 , a multi-modal intelligent baby stroller system based on SLAM navigation includes: a mobile chassis (1), a scissor lift device (2), a rocking device (3), a tilting device (4), a light-shielding device (5), a camera (22), and an autonomous navigation and intelligent voice interaction system.

[0041] The mobile chassis (1) includes a braking device (11), a chassis housing (12), a threaded post (6), a bottom plate (13), a cover plate (14), and a control core component. The braking device (11) is connected to the driving wheels (15) and can be forced to brake by stepping on it with the foot. The depth camera (16), the temperature and humidity sensor (17), and the air quality sensor (18) are integrated on the chassis housing for perceiving the surrounding environment. The upper end of the bottom plate (13) places the control core component of the baby stroller for the main navigation and intelligent voice interaction of the baby stroller.

[0042] The control core component includes an Arduino core board, an industrial control computer, an ESP8266 module, a lidar, an IMU, an intelligent voice interaction module, a servo motor driver, a servo motor (19), and driving wheels (15). The industrial control computer and the ESP8266 module are integrated on the bottom plate (13) for processing the image information collected by the camera (22) and sending it to the parents. The servo motor driver and the encoder are integrated on the bottom plate (13) for driving the chassis servo motor (19). The lidar, the IMU, and the intelligent voice interaction module are respectively placed on the cover plate (14) for constructing a multi-scale 3D environment map, path planning, and voice control.

[0043] By cooperating the control core component with the mobile chassis (1), autonomous navigation and obstacle avoidance of the baby stroller can be achieved. A three-dimensional point cloud map is formed by the laser beam emitted by the lidar, and then combined with the angular velocity and acceleration data of the IMU to achieve baby stroller positioning and construction of a multi-scale 3D environment map, supporting the mobile chassis to autonomously avoid obstacles and plan the optimal path in an unknown environment. The movement control is achieved by the Arduino core board controlling the servo motor (19). Combining the real-time feedback of the surrounding environment information and the baby stroller attitude position data, the rotation speed of the servo motor (19) is dynamically adjusted to achieve differential control.

[0044] The intelligent voice interaction module cooperates with the mobile chassis (1) to realize voice control of the baby stroller to move to a specified position. First, the front end of the voice signal is processed. The audio signal is analyzed through voice activity detection (VAD), and at the same time, an adaptive filtering algorithm is used to eliminate environmental noise and echo. Then, voice wake-up and command recognition are carried out. A specific wake-up word (such as "Hello, Xiaoai") is listened to in real time. After successful wake-up, automatic speech recognition (ASR) is used to convert the voice into text and then transfer it to the dialogue management module for interaction. Finally, dialogue management and response generation are carried out. Through the context interaction mode, reasonable dialogue text is generated, and then using the TTS speech synthesis technology, the obtained text is converted into natural speech and output to the parents. This system also supports the selection of emotional voices to enhance the affinity and realizes a natural and smooth voice interaction experience.

[0045] The camera (22) cooperates with the industrial control computer and the ESP8266 module to realize real-time monitoring of the baby's state and remote alarm. When the camera (22) detects abnormal behaviors, such as risk behaviors like the baby turning over or head occlusion, the industrial control computer performs image analysis. After determining the abnormality, the information is sent to the parent's mobile phone APP through the ESP8266 module. Moreover, the industrial control computer can identify the baby's crying by running an AI audio analysis model. If it exceeds the set decibel threshold, the shaking device (3) and the music playing mode are activated to simulate natural shaking, and at the same time, a mobile phone push alarm message is triggered.

[0046] The lifting device (2) includes a lower load-bearing platform (25), a middle load-bearing platform (26), a bearing seat (27), a linear guide rail slider (28), a scissor lift module (7), a ball screw nut module (30), and a stepping motor 1 (31). The scissor lift module (7) is connected to the lower load-bearing platform (25) and the middle load-bearing platform (26) through the bearing seat (27). The ball screw nut module (30) is driven by the stepping motor 1 (31) to drive the scissor lift module (7) to expand or contract. The safety seat (10) smoothly rises and falls along the linear guide rail slider (29) to achieve height adjustment. The lower load-bearing platform (25) and the bearing seat provide stable support. This device adopts a multi-dimensional adjustment function, supports a flexible lifting structure, meets the usage requirements in different scenarios, can easily adjust the seat height according to actual needs, is convenient for picking up or putting down the baby, and improves the convenience of use.

[0047] The rocking device (3) includes a middle load-bearing platform (26), an upper load-bearing platform (32), a rotating mechanism (9), a compression spring (8), and a spring limit block (33). The rotating mechanism is driven by a stepping motor 2 (34) and drives the rotating mechanism (9) to perform a rotating motion through a 2gt synchronous belt (35). The compression spring (8) provides buffering and reset forces. The two cooperate to achieve the rocking function of the safety seat (10). At the same time, by changing the angle of the rotating mechanism (9), different rocking modes can be selected, and a music playback mode is equipped to simulate natural rocking to provide comfortable soothing for the baby. When the baby cries, the rocking function is started in combination with soothing music to help the baby fall asleep calmly.

[0048] The forward tilt device (4) includes an upper load-bearing platform (32), an electric push rod (36), a fixed pin, a bearing block 2 (37), a smooth shaft 3 (38), and a safety seat (10). The electric push rod (36) moves up and down, and the safety seat (10) rotates around the smooth shaft 3 (38) to achieve the forward tilt function. The seat angle is automatically / manually adjusted according to different scenarios (such as sleeping, feeding) to optimize the riding stability and comfort, meet the needs of different scenarios such as when the baby is awake and eating, and ensure the comfort and convenience of the baby.

[0049] The light-shielding device (5) includes a safety seat (10), a light-shielding cover (39), a rotating motor (40), and a light sensor (41). The light-shielding cover (39) is arc-shaped. When the light sensor (41) detects that the sunlight is too strong and directly shines on the safety seat (10), the rotating motor (40) is started, and the light-shielding cover (39) rotates to achieve automatic sunshading, providing guarantee for outdoor travel. When the outdoor sunlight is relatively dazzling, the light-shielding cover automatically closes to avoid burning the baby's skin.

[0050] This invention patent has important significance and purpose in the fields of structural optimization, artificial intelligence technology, SLAM navigation technology, and deep learning. First of all, the present invention optimizes the mechanical device through intelligent control technology, adapts to different scenario requirements with the lifting device, dynamically adjusts the vehicle body posture, and the light-shielding, forward tilt, and rocking devices automatically adjust the light-shielding cover angle, forward tilt angle, and rocking frequency, ensuring the baby's sleep state and the comfort of travel.

[0051] Moreover, to enhance the safety of travel through multi-modal perception fusion, multi-sensor fusion such as lidar, depth camera, and IMU is adopted to construct a high-precision environmental model around the stroller in real time. Combining SLAM technology to achieve dynamic obstacle avoidance, path planning, and intelligent following, it improves travel convenience. In public places, parents don't need to tightly hold the push handle, and the stroller can automatically follow and avoid obstacles, making parents more relaxed. At the same time, it is also equipped with an intelligent monitoring system, autonomous navigation, and intelligent voice interaction system, supporting voice control, remote monitoring, and real-time feedback to enhance parent-child interaction. Parents can view the baby's status at any time through the mobile APP and keep interacting with the baby even when they are busy.

[0052] This invention patent innovatively combines artificial intelligence, mechanical engineering, and human-machine design to create a modern stroller that integrates safety, intelligence, and comfort, providing an efficient solution for the parenting scenario and having significant market application potential and social value.

[0053] Embodiment 2:

[0054] A SLAM navigation method for an intelligent stroller, at least including:

[0055] S1: First, the lidar provides high-precision point cloud data, the depth camera supplements texture information, and the IMU obtains the angular velocity and acceleration data of the stroller. The three work together to obtain the perception data of the surrounding environment;

[0056] S2: Adopt adaptive feature point extraction technology to dynamically extract edge points (such as ORB features) and plane points (such as the ground, walls) to ensure the safety of the stroller's passing area;

[0057] S3: Use the perception data for self-localization to determine the position and attitude of the stroller in the environment;

[0058] S4: According to the environmental perception data and the result of self-localization, convert the perception data into a representation of the environment, and then combine the visual ORB feature points to construct a multi-scale 3D environmental map, laying the foundation for the next navigation and path planning;

[0059] S5: Through factor graph optimization (such as GTSAM), fuse new perception data in real time and eliminate the interference of dynamic objects, and update the map information in a timely manner to ensure the real-time nature and accuracy of the 3D environmental map.

[0060] S6: Introduce dynamic weights and heuristic functions to optimize the A* algorithm for dynamic path planning, use Kalman filtering to distinguish dynamic interference points, detect dynamic obstacle avoidance objects in real time to filter out interference points, and adjust the path in real time to ensure the safety of the stroller's travel.

[0061] The above description is not intended to impose any form of limitation on the present invention. Although the present invention has been disclosed through the above embodiments, it is not intended to limit the present invention. Any person skilled in the relevant art can make some changes or modifications within the scope of the technical solution of the present invention to obtain equivalent embodiments with equivalent changes. However, as long as the content does not deviate from the technical solution of the present invention, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A multi-modal intelligent baby stroller system and control method based on SLAM navigation, the components of which include: Mobile chassis (1), lifting device (2), shaking device (3), forward tilting device (4), shading device (5), intelligent monitoring system, autonomous navigation and intelligent voice interaction system; The mobile chassis (1) is connected to the lifting device (2) of the baby carriage via four threaded columns (6); The shaking device (3) is supported by a scissor-type lifting module (7) and is connected to the lifting device (2); The forward tilting device (4) is connected to the shaking device (3) by means of four compression springs (8); The shading device (5) is connected to the safety seat (10) via a rotating mechanism (9).

2. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 1, characterized in that: The mobile chassis (1) comprises a brake device (11), a chassis shell (12), a threaded column (6), a base plate (13), a cover plate (14), and a control core component. The brake device (11) is connected to a driving wheel (15). A depth camera (16), a temperature and humidity sensor (17), and an air quality sensor (18) are integrated on the chassis shell (12). The baby carriage control core component is placed on the upper end of the base plate (13).

3. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 2, characterized in that: The control core component comprises an Arduino core board, a servo motor driver and an encoder, a servo motor (19), a driving wheel (15), and a universal wheel (20). The Arduino core board is placed on a base plate (13). The servo motor driver and the encoder are integrated around the Arduino core board. The servo motor (19) is fixed to a motor bracket (21) by bolts. The motor bracket (21) is connected to a mounting hole of the base plate (13) and is symmetrically distributed at two ends below the base plate (13). The servo motor (19) is fixed to the driving wheel (15) by a shaft sleeve.

4. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 1, characterized in that: The intelligent monitoring system comprises a camera (22), an industrial control computer, and an ESP8266. The camera (22) is fixed above the safety seat (10) to collect image information. The industrial control computer compares and analyzes the image information. The ESP8266 sends information to a mobile phone / PC to detect whether the baby rolls over or has his head blocked. If there is any situation, an alarm is issued in time and information is sent.

5. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 1, characterized in that: The autonomous navigation and intelligent voice interaction system for the baby carriage comprises a SLAM algorithm, a path planning algorithm and various hardware modules, wherein the hardware modules include a depth camera (16), a laser radar (23), an IMU, a voice sensor (24), a temperature and humidity sensor (17), and an air quality sensor (18). The depth camera (17) is fixed at the front end of the chassis shell (12), the laser radar (23) is fixed at the edge of the cover plate (14), the laser radar sensor is connected to the laser radar (23) and fixed on the cover plate (14), the voice sensor (24) and the IMU are fixed on the bottom plate (13), and the temperature and humidity sensor (17) and the air quality sensor (18) are fixed on the side of the chassis shell (12).

6. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 1, characterized in that: The lifting device (2) comprises a lower load-bearing platform (25), a middle load-bearing platform (26), a bearing seat 1 (27), a linear guide rail (28), a linear guide rail slider (29), a scissors-type lifting module (7), a ball screw nut module (30), and a stepper motor 1 (31). The scissors-type lifting module (7) is connected to the lower load-bearing platform (25) and the middle load-bearing platform (26) through the bearing seat (27). The ball screw nut module (30) is driven by the stepper motor 1 (31) to drive the scissors-type lifting module (7) to expand or contract. The safety seat (10) is lifted and lowered smoothly along the linear guide rail slider (29) to achieve height adjustment. The lower load-bearing platform (25) and the bearing seat (27) provide stable support.

7. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 1, characterized in that: The rocking device (3) comprises a middle load-bearing platform (26), an upper load-bearing platform (32), a rotating mechanism (9), a compression spring (8), and a spring stopper (33). The rotating mechanism (9) is driven by a stepper motor 2 (34), and the rotating mechanism (9) is driven to rotate via a 2gt synchronous belt (35). The compression spring (8) provides a buffer and a reset force. The two cooperate to realize the rocking function of the safety seat (10). At the same time, by changing the angle of the rotating mechanism (9), different rocking modes can be selected.

8. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 1, characterized in that: The forward tilting device (4) comprises an upper load-bearing platform (32), an electric push rod (36), a fixing pin, a bearing seat 2 (37), an optical axis 3 (38), and a safety seat (10); the electric push rod (36) moves up and down, and the safety seat (10) rotates around the optical axis 3 (38), thereby realizing a forward tilting function.

9. The multi-modal intelligent baby stroller system and control method based on SLAM navigation according to claim 1, characterized in that: The shading device (5) comprises a safety seat (10), a shading cover (39), a rotating motor (40), and a light sensor (41); the shading cover (39) is arc-shaped; when the light sensor (41) detects that the sunlight is too strong and directly shines on the safety seat (10), the rotating motor (40) is started and the shading cover (39) is rotated, thereby achieving automatic shading.

10. A SLAM navigation method for a smart baby stroller according to claim 5, comprising: S1: First, the stroller obtains perception data of the surrounding environment through the sensors it carries (lidar, depth camera, IMU); S2: Adopting adaptive feature point extraction technology, edge points and plane points are dynamically extracted from the acquired image; S3: Use the perception data to locate the stroller and determine its position and posture in the environment; S4: Based on the environmental perception data and the results of self-positioning, the perception data is converted into a representation of the environment, and then combined with visual ORB feature points to build a multi-scale 3D environmental map; S5: Use perception data to update map information in a timely manner to ensure the real-time and accuracy of the 3D environment map; S6: Introduce the improved A* algorithm for dynamic path planning, real-time detection of dynamic obstacle avoidance and filtering of interference points, and real-time adjustment of the path to ensure the safety of baby strollers.

Images