Intelligent baby carriage control method with active obstacle avoidance function

Abstract

The invention discloses an intelligent baby carriage control method and system with an active obstacle avoidance function. The method comprises the following steps: acquiring multi-mode environment sensing data and extracting features, and acquiring a ground negative height change area; collecting real-time motion state data, and constructing a wheel advancing projection area in a future preset time period; calculating geometric overlapping parameters of the projection area and the negative height change area; and if the overlapping parameter meets the preset risk condition, generating a braking or bypassing instruction. According to the method, negative obstacles are accurately recognized through multi-modal fusion, geometric overlapping parameter quantification risks are introduced, the problems that in the prior art, the sensing dimension is single, steps cannot be recognized, and the false alarm rate is high are solved, crossing from passive response to active track pre-judgment is achieved, and the safety of the intelligent baby carriage is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control and robot navigation technology, specifically to a control method for an intelligent children's vehicle with active obstacle avoidance function. Background Technology

[0002] With the development of smart technology, smart strollers are becoming increasingly popular. However, when pushing a stroller, parents' view is often obstructed or distracted, increasing the risk of accidents. Most current smart strollers are equipped with ultrasonic radar or infrared sensors for obstacle avoidance.

[0003] However, existing technologies have significant drawbacks:

[0004] (1) Single perception dimension: Traditional ultrasonic or infrared sensors are mainly effective against protruding obstacles and are difficult to identify negative height change areas such as ground pits, steps or platform gaps (i.e. negative obstacles).

[0005] (2) High false alarm rate. In crowded or narrow environments, the judgment logic based solely on distance thresholds is prone to frequent sudden stops, affecting the implementation experience.

[0006] (3) The response is delayed, and it is often too late to brake when the danger is detected.

[0007] Currently, there is a lack of a good technology that can overcome the shortcomings of existing technologies.

[0008] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0009] This application provides a smart stroller control method and system with active obstacle avoidance function, aiming to solve the problems of existing technologies being unable to effectively identify negative obstacles and having a high false alarm rate in obstacle avoidance control.

[0010] The technical solution of the present invention is as follows:

[0011] On one hand, the present invention provides a control method for an intelligent children's vehicle with active obstacle avoidance function. The method includes the following steps: acquiring multimodal environmental perception data of the intelligent children's vehicle's direction of travel, and extracting features from the multimodal environmental perception data to obtain a negative height change area on the ground; collecting real-time motion state data of the intelligent children's vehicle, and constructing a wheel projection area of ​​the intelligent children's vehicle within a future preset time period based on the real-time motion state data; calculating the geometric overlap parameter between the wheel projection area and the negative height change area; determining whether the geometric overlap parameter meets a preset risk condition; if it does, generating a corresponding obstacle avoidance control command based on the geometric overlap parameter to control the intelligent children's vehicle to perform braking or detour operations.

[0012] Optionally, the step of acquiring multimodal environmental perception data of the intelligent stroller's direction of travel and extracting features from the multimodal environmental perception data to obtain the negative height change region on the ground specifically includes: simultaneously acquiring point cloud data from a lidar and image data from a camera; mapping the point cloud data and the image data to a unified vehicle coordinate system using a joint calibration matrix to obtain spatiotemporally aligned fused data; calculating the ground elevation gradient and depth difference based on the fused data, and extracting the boundary contour that meets the preset negative feature conditions as the negative height change region.

[0013] Optionally, extracting the boundary contour that satisfies the preset negative feature conditions specifically includes: constructing a local ground elevation map and traversing and calculating the height gradient of the point set in the neighborhood; if the depth difference between the current point and the neighborhood points is detected to be greater than a preset depth threshold, and the height gradient is greater than a preset gradient threshold, then the point is determined to be a candidate boundary point of the negative height change region; and the candidate boundary point is fitted using an edge detection algorithm to obtain a continuous boundary contour curve.

[0014] Optionally, the step of collecting the real-time motion state data of the smart children's vehicle specifically includes: establishing a motion state vector of the smart children's vehicle, the motion state vector including at least position parameters and velocity parameters; constructing the state transition equation and observation equation of the discrete-time system; inputting the motion state vector into a Kalman filter model for time update and measurement update, and outputting the optimal estimated state as the real-time motion state data.

[0015] Optionally, calculating the geometric overlap parameter between the wheel's projected travel area and the negative height change area specifically includes: calculating the wheel's trajectory boundary within a future preset time period based on the axle width, current steering angle, and real-time motion state data of the smart stroller; calculating the geometric intersection area between the area enclosed by the trajectory boundary and the negative height change area; and determining the ratio of the geometric intersection area to the total area swept by the wheel as the geometric overlap parameter.

[0016] Optionally, determining whether the geometric overlap parameter meets preset risk conditions; if it does, generating a corresponding obstacle avoidance control command based on the geometric overlap parameter, specifically including: calculating the relative approach time of the smart stroller to the boundary of the negative height change area; if the geometric overlap parameter is greater than zero, classifying the risk level based on the relative approach time; if the risk level is level one risk, outputting a warning signal; if the risk level is level two risk, outputting a deceleration braking command; if the risk level is level three risk, outputting an emergency stop braking command or a path detour command.

[0017] Optionally, if the risk level is level two, a deceleration braking command is output, specifically including: constructing a dynamic desired speed model, and using the geometric overlap parameters and the current distance between the smart stroller and the negative height change area as model inputs to obtain the dynamic desired speed; calculating the speed deviation value between the current speed of the smart stroller and the dynamic desired speed; inputting the speed deviation value into a PID controller for proportional, integral, and derivative operations, and outputting a braking signal strength matching the dynamic speed.

[0018] Optionally, if the risk level is level three, an emergency stop braking command or a detour command is output, specifically including: during the braking operation, real-time acquisition of feedback data from wheel speed sensors and calculation of the expected stopping distance of the smart children's vehicle; determining whether the expected stopping distance is greater than the current distance between the smart children's vehicle and the negative height change area; if it is greater, determining that the braking operation cannot complete the avoidance, and triggering the detour command, and controlling the steering of the smart children's vehicle according to the optimized travel direction adjustment signal.

[0019] Optionally, controlling the steering of the smart stroller according to the optimized travel direction adjustment signal specifically includes: constructing a virtual potential field model containing target gravitational field parameters and negative obstacle repulsive field parameters; determining whether the distance between the smart stroller and the negative height change region is less than a preset repulsive influence range; if it is less, generating a repulsive vector based on the negative obstacle repulsive field parameters, and combining it with the gravitational vector generated based on the target gravitational field parameters to synthesize the optimized travel direction adjustment signal.

[0020] On the other hand, the present invention also provides an intelligent stroller control system with active obstacle avoidance function. The system includes: a multimodal perception module for acquiring multimodal environmental perception data of the stroller's direction of travel and extracting features from the multimodal environmental perception data to obtain the negative height change area on the ground; a state estimation module for collecting real-time motion state data of the stroller and constructing the wheel projection area of ​​the stroller in a future preset time period based on the real-time motion state data; a risk calculation module for calculating the geometric overlap parameters between the wheel projection area and the negative height change area; and an obstacle avoidance control module for determining whether the geometric overlap parameters meet preset risk conditions. If they do, a corresponding obstacle avoidance control command is generated based on the geometric overlap parameters to control the stroller to perform braking or detour operations.

[0021] The beneficial effects of this invention are as follows:

[0022] This application achieves accurate detection of negative height variation areas such as ground depressions and step gaps by acquiring multimodal fusion data from lidar and cameras, and using gradient calculation and edge detection algorithms to extract features from the multimodal environmental perception data. This overcomes the limitation of traditional ultrasonic or infrared sensors that can only detect protruding obstacles, filling the gap in the current technology for sensing negative ground features.

[0023] Meanwhile, the invention abandons the triggering logic based solely on absolute distance thresholds and innovatively introduces geometric overlap parameters as the core criterion. By constructing the wheel's projected travel area over a preset time period and calculating its geometric intersection with the danger zone, the system can intelligently distinguish between safe close-range passage and substantial fall risk. This solves the problem of false alarms caused by simply being too close in crowded or narrow environments, reduces the false alarm rate, and ensures smooth operation.

[0024] Furthermore, this invention achieves posterior estimation and prospective prediction of the smart children's vehicle's motion state by collecting real-time motion state data and inputting it into a Kalman filter model. The system does not trigger only upon touching a dangerous boundary; instead, it anticipates potential overlap risks based on the predicted wheel travel projection area and dynamically adjusts the braking timing based on relative proximity. This prediction-based active defense mechanism overcomes the response lag defects caused by sensor sampling delays or data processing time in existing technologies, ensuring the timeliness and effectiveness of braking. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the method flow provided in Embodiment 1 of this application; Figure 2 This is a schematic diagram illustrating the spatiotemporal alignment principle of multimodal data in Embodiment 1 of this application. Figure 3 This is a schematic diagram illustrating the calculation of the overlap between the wheel travel projection and the negative obstacle in Embodiment 1 of this application; Figure 4 This is a schematic diagram of the system module structure provided in Embodiment 3 of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0027] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0028] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0029] As mentioned earlier, most current smart strollers are equipped with ultrasonic radar or infrared sensors for obstacle avoidance. This obstacle avoidance method has significant drawbacks: First, the perception dimension is limited. Traditional ultrasonic or infrared sensors are mainly effective against raised obstacles and have difficulty identifying areas with negative height changes (i.e., negative obstacles) such as ground pits, steps, or platform gaps. Second, the false alarm rate is high. In crowded or narrow environments, the judgment logic based solely on distance thresholds can easily lead to frequent sudden stops, affecting the strolling experience. Third, the response is delayed. Often, by the time danger is detected, it is too late to brake.

[0030] Currently, there is a lack of a good technology that can overcome the shortcomings of existing technologies.

[0031] To address this, the present invention provides an intelligent stroller control method with active obstacle avoidance function, which effectively solves the defects in the prior art. The present invention solves this problem in the following way.

[0032] Example 1:

[0033] Please see Figures 1 to 3 This embodiment details the operation of the intelligent stroller control method with active obstacle avoidance function, specifically including the following steps:

[0034] Step S100: Obtain multimodal environmental perception data of the direction of travel of the smart stroller, and extract features from the multimodal environmental perception data to obtain the negative height change area on the ground.

[0035] Specifically, acquiring multimodal environmental perception data includes simultaneously acquiring point cloud data from lidar. and camera image data To achieve data fusion, a joint calibration matrix is ​​used to map the point cloud data and the image data to a unified vehicle coordinate system. The rotation matrix from the LiDAR coordinate system to the camera coordinate system is set as follows: Translation vector Then any point in the point cloud Coordinates mapped to the image plane Calculated using the following formula:

[0036]

[0037] in, Indicates the scaling factor; Represents the pixel coordinates on the image plane; Represents the camera intrinsic parameter matrix; This represents the rotation matrix from the lidar coordinate system to the camera coordinate system; This represents the translation vector from the lidar coordinate system to the camera coordinate system; This represents the three-dimensional coordinates of a point in the lidar point cloud. The spatiotemporally aligned fused data is obtained using the above formula.

[0038] Furthermore, the ground elevation gradient and depth difference are calculated based on the fused data. A local ground elevation map is constructed, and the set of points in the neighborhood is calculated iteratively. Height gradient :

[0039]

[0040] in, Indicates coordinates The height gradient magnitude at that location; Indicates the elevation value of the ground; Indicates elevation at Partial derivatives in the direction; Indicates elevation at Partial derivatives in the direction.

[0041] If the current point height is detected Average height of neighboring points The depth difference is greater than the preset depth threshold. And the height gradient Greater than the preset gradient threshold If the point is found to be a candidate boundary point for a region of negative height change, then the following condition must be met:

[0042]

[0043] Finally, edge detection algorithms (such as the Canny algorithm) are used to fit the candidate boundary points to obtain a continuous boundary contour curve. As the negative height change region .

[0044] Step S200: Collect real-time motion status data of the smart stroller, and construct the wheel travel projection area of ​​the smart stroller within a future preset time period based on the real-time motion status data.

[0045] To improve the accuracy of state estimation, this step employs a Kalman filter model. The motion state vector of the intelligent children's vehicle is then established. ,in For position parameters, Let the velocity parameter be used. Construct the state transition equation for the discrete-time system. and observation equations The motion state vector is input into a Kalman filter model for time and measurement updates, and the optimal estimated state is output. (i.e., the real-time motion state data):

[0046]

[0047] in, express The prior state estimate at time t; Represents the state transition matrix; express The posterior optimal state estimate at time t; Represents the prior error covariance matrix; Represents the posterior error covariance matrix; Represents the process noise covariance matrix; Indicates Kalman gain; Represents the observation matrix; Represents the observation noise covariance matrix; express The actual observation vector at any given time (i.e., the sensor measurement value); express The posterior optimal state estimate after the time-time update.

[0048] Subsequently, based on the real-time motion data, the trajectory boundary of the wheel within a preset time period is calculated. This is based on the width of the children's wheel axle. and current steering angle The coordinates of the left and right contact points of the front wheels of the vehicle body change with time. The equation of the projected trajectory is:

[0049]

[0050] in, express Predicted coordinates of the center position of the front wheel of the vehicle at any given moment; This indicates the initial position of the vehicle at the current moment; This represents the optimal estimated speed of the Kalman filter output; Indicates the heading angle of the vehicle at the current moment; This indicates the front and rear wheelbase of the smart stroller; Indicates the current front wheel steering angle; This represents the time variable for integration.

[0051] Step S300: Calculate the geometric overlap parameters between the wheel travel projection area and the negative height change area.

[0052] Specifically, the wheel travel projection area is calculated. The negative height change region identified in step S100 Geometric intersection area The geometric intersection area is compared with the total area swept by the wheel. The ratio of these values ​​is determined as the geometric overlap parameter. :

[0053]

[0054] in, Indicates the geometric overlap parameter; This represents the geometric intersection area of ​​the wheel's projected travel area and the area of ​​negative height change. This represents the total area of ​​the wheel's projected area (i.e., the area of ​​the road surface swept by the wheel within the predicted time).

[0055] Step S400: Determine whether the geometric overlap parameters meet the preset risk conditions; if they do, generate the corresponding obstacle avoidance control command based on the geometric overlap parameters.

[0056] First, calculate the relative approach time of the smart stroller to the boundary of the negative height change region. If the geometric overlap parameter The risk level is then determined based on the relative proximity of the time.

[0057] like and If the risk level is determined to be Level 1, a warning signal will be output.

[0058] like and The risk level is determined to be Level 2, and a deceleration and braking command is output.

[0059] like and If the risk level is determined to be Level 3, an emergency stop command or a detour command will be issued.

[0060] When outputting deceleration and braking commands, a dynamic desired speed model is constructed. And calculate the deviation. The speed deviation value is input into a PID controller for proportional, integral, and derivative calculations, and the output braking signal strength is then determined. :

[0061]

[0062] in, express The strength of the braking control signal output at all times; These represent the proportional coefficient, integral coefficient, and differential coefficient, respectively. This represents the deviation between the current speed and the dynamically desired speed (i.e., ); Representing the integral variable

[0063] If a path detour is deemed necessary (e.g., when simple braking cannot avoid a collision), an artificial potential field method is introduced. This involves constructing a potential field containing the target gravitational field parameters. and negative obstacle repulsive field parameters The virtual potential field model. The negative obstacle repulsive force field function is defined as:

[0064]

[0065] in, The function representing the repulsive potential field generated by a negative obstacle; Indicates the repulsive gain coefficient (positive number); Indicates the current location of the smart stroller Boundary of negative height change region The shortest Euclidean distance between them; This indicates the preset repulsive force influence range distance threshold.

[0066] Calculate the resultant force The optimized travel direction adjustment signal is synthesized to control the steering of the stroller.

[0067] Example 2:

[0068] Based on the above embodiments, in order to further clarify and completely explain the technical solutions therein, the present invention also provides Embodiment Two. Embodiment Two is based on the method steps described in Embodiment One, and verifies the operational logic of the method in a real-world scenario by setting specific example data. To verify the effectiveness of this method, the following application scenario is constructed:

[0069] Smart strollers are becoming increasingly popular. Traveling at a speed along a straight line straight ahead, the distance ahead is... There is a downward-sloping step (negative obstacle) at this location, with a height difference of [missing information]. Set a preset time period. Children's wheel axle width .

[0070] First, step S100 is executed to acquire multimodal environmental perception data of the smart stroller's direction of travel, and feature extraction is performed on the multimodal environmental perception data. The lidar collects point cloud data from the front. The point cloud is mapped to the vehicle coordinate system using a calibration matrix. Then, select the front of the vehicle. sampling points at the location The system reads the height of that point. (Road surface level), and read the average height of its neighboring points (below the step). The depth difference was calculated. And the height gradient at that location Much greater than the value for a flat road surface. This is because the calculated depth difference exceeds the preset depth threshold. And the gradient is greater than the preset gradient threshold. The system determines the coordinates. The boundary of the negative height change region is defined at this point, and the boundary contour curve that transversely penetrates the road surface is extracted. .

[0071] Subsequently, in step S200, real-time motion status data of the smart stroller is collected, and a wheel travel projection area is constructed. This addresses the issue of fluctuations in the observation data read by the wheel speed sensor (e.g., ...). In the case of [condition], the observed data is input into the Kalman filter model, and after time updates and measurement updates, the optimal estimated state is output: position. ,speed heading angle Based on the preset time period And the optimal estimated speed, predict the travel distance Combined with wheel and axle width Construct the wheel travel projection area In this scenario, the wheel's projection area is a covered coordinate range. A rectangular area.

[0072] Next, step S300 is executed to calculate the geometric overlap parameters. Since the wheel's travel projection area extends to... Crossing the area located Calculate the geometric intersection area at the step boundary. The rectangular portion from the edge of the step to the predicted endpoint, i.e. Simultaneously calculate the total area swept by the wheel. The system determines the geometric overlap parameters. .

[0073] Finally, according to step S400, the risk is assessed and instructions are generated. Because... If the preset risk conditions are met, the system further calculates the relative proximity time. The relative proximity time is compared with a preset safety threshold. and emergency braking threshold Comparison, because The current risk level is determined to be Level 2. The system generates a deceleration and braking command and constructs a dynamic desired speed. and the calculated speed deviation Input the PID controller and output the corresponding braking voltage signal to control the motor to perform flexible deceleration, thereby preventing the stroller from falling off the steps.

[0074] Example 3:

[0075] Based on the same general inventive concept, and in order to provide a clearer and more complete explanation of the technical solutions in the foregoing embodiments, the present invention also provides Embodiment Three. Please refer to... Figure 4 This third embodiment provides an intelligent stroller control system with active obstacle avoidance function. The system includes the following modules that are physically connected and communicate with each other:

[0076] The multimodal perception module is used to acquire multimodal environmental perception data of the direction of travel of the smart stroller, and to extract features from the multimodal environmental perception data to obtain the negative height change area on the ground.

[0077] The state estimation module is used to collect real-time motion state data of the smart stroller and construct the wheel travel projection area of ​​the smart stroller within a preset time period based on the real-time motion state data.

[0078] The risk calculation module is used to calculate the geometric overlap parameters between the wheel travel projection area and the negative height change area;

[0079] The obstacle avoidance control module is used to determine whether the geometric overlap parameters meet the preset risk conditions; if they do, it generates a corresponding obstacle avoidance control command based on the geometric overlap parameters to control the smart stroller to perform braking or detour operations.

[0080] It should be understood that, in the embodiments of the present invention, "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0081] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0082] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0083] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, apparatuses, or units, or they may be electrical, mechanical, or other forms of connection.

[0084] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0085] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0086] From the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented in hardware, firmware, or a combination thereof. When implemented in software, the above-described functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media can be any available medium accessible to a computer. For example, but not limited to, computer-readable media can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer. Furthermore, any connection can suitably be a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of the medium. As used in this invention, disk and disc include compressed optical discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, wherein disks typically magnetically copy data, while discs optically copy data using lasers. The combinations described above should also be included within the scope of protection for computer-readable media.

[0087] In summary, the above description is merely a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A control method for an intelligent children's vehicle with active obstacle avoidance function, characterized in that, The method includes: Acquire multimodal environmental perception data of the direction of travel of the intelligent stroller, and extract features from the multimodal environmental perception data to obtain the negative height change area on the ground; Collect real-time motion status data of the smart stroller, and construct the wheel travel projection area of ​​the smart stroller within a future preset time period based on the real-time motion status data; Calculate the geometric overlap parameters between the wheel travel projection area and the negative height change area; Determine whether the geometric overlap parameters meet the preset risk conditions; if they do, generate corresponding obstacle avoidance control commands based on the geometric overlap parameters to control the smart stroller to perform braking or detour operations.

2. The method according to claim 1, characterized in that, The process of acquiring multimodal environmental perception data of the intelligent stroller's direction of travel and extracting features from the multimodal environmental perception data to obtain negative height change areas on the ground specifically includes: Simultaneously acquire point cloud data from lidar and image data from cameras; The point cloud data and the image data are mapped to a unified vehicle coordinate system using a joint calibration matrix to obtain spatiotemporally aligned fused data; Based on the fused data, the ground elevation gradient and depth difference are calculated, and the boundary contours that meet the preset negative feature conditions are extracted as the negative height change region.

3. The method according to claim 2, characterized in that, The extraction of the boundary contour that satisfies the preset negative feature condition specifically includes: Construct a local ground elevation map and iterate through the points in the neighborhood to calculate the height gradient. If the depth difference between the current point and its neighboring points is greater than a preset depth threshold, and the height gradient is greater than a preset gradient threshold, then the point is determined to be a candidate boundary point of a negative height change region. The boundary candidate points are fitted using an edge detection algorithm to obtain a continuous boundary contour curve.

4. The method according to claim 1, characterized in that, The collection of real-time motion status data of the smart stroller specifically includes: Establish the motion state vector of the intelligent children's vehicle, wherein the motion state vector includes at least position parameters and velocity parameters; Construct the state transition equations and observation equations for the discrete-time system; The motion state vector is input into the Kalman filter model for time and measurement updates, and the optimal estimated state is output as the real-time motion state data.

5. The method according to claim 1, characterized in that, The calculation of the geometric overlap parameters between the wheel travel projection area and the negative height change area specifically includes: Based on the wheel axle width, current steering angle, and real-time motion status data of the intelligent children's vehicle, the boundary of the wheel's trajectory within a preset time period is calculated. Calculate the area of ​​the geometric intersection between the region enclosed by the boundary of the travel trajectory and the region of negative height change; The ratio of the geometric intersection area to the total area swept by the wheel is determined as the geometric overlap parameter.

6. The method according to claim 1, characterized in that, The determination is made as to whether the geometric overlap parameter meets the preset risk conditions; If satisfied, then a corresponding obstacle avoidance control command is generated based on the geometric overlap parameters, specifically including: Calculate the relative approach time of the intelligent stroller to the boundary of the negative height change region; If the geometric overlap parameter is greater than zero, the risk level is determined based on the relative proximity time. If the risk level is Level 1, a warning signal will be output. If the risk level is level two, then a deceleration and braking command will be output. If the risk level is level three, then an emergency stop command or a detour command will be output.

7. The method according to claim 6, characterized in that, If the risk level is level two, then a deceleration and braking command will be output, specifically including: A dynamic expected speed model is constructed, and the geometric overlap parameters and the current distance between the smart stroller and the negative height change region are used as model inputs to obtain the dynamic expected speed. Calculate the speed deviation between the current speed of the smart stroller and the dynamically desired speed; The speed deviation value is input into the PID controller for proportional, integral, and derivative operations, and the output is a braking signal strength that matches the dynamic speed.

8. The method according to claim 6, characterized in that, If the risk level is level three, then an emergency stop command or a detour command will be output, specifically including: During the braking operation, feedback data from the wheel speed sensor is acquired in real time, and the estimated stopping distance of the smart stroller is calculated. Determine whether the expected stopping distance is greater than the current distance between the smart stroller and the negative height change area; If the value is greater than the value, it is determined that the braking operation cannot complete the avoidance and the path detour instruction is triggered. The intelligent children's vehicle is then controlled to turn according to the optimized travel direction adjustment signal.

9. The method according to claim 8, characterized in that, The step of controlling the steering of the intelligent stroller based on the optimized direction of travel adjustment signal specifically includes: Construct a virtual potential field model that includes the target gravitational field parameters and the negative obstacle repulsive field parameters; Determine whether the distance between the smart stroller and the negative height change area is less than the preset repulsive force influence range; If it is less than the target gravitational field parameter, a repulsion vector is generated based on the negative obstacle repulsion field parameter, and combined with the gravitational vector generated based on the target gravitational field parameter to synthesize the optimized travel direction adjustment signal.

10. A smart stroller control system with active obstacle avoidance function, characterized in that, The system includes: The multimodal perception module is used to acquire multimodal environmental perception data of the direction of travel of the smart stroller, and to extract features from the multimodal environmental perception data to obtain the negative height change area on the ground. The state estimation module is used to collect real-time motion state data of the smart stroller and construct the wheel travel projection area of ​​the smart stroller within a preset time period based on the real-time motion state data. The risk calculation module is used to calculate the geometric overlap parameters between the wheel travel projection area and the negative height change area; The obstacle avoidance control module is used to determine whether the geometric overlap parameters meet the preset risk conditions; if they do, it generates a corresponding obstacle avoidance control command based on the geometric overlap parameters to control the smart stroller to perform braking or detour operations.

Images