Method and system for controlling reversing balance of two-wheeled vehicle by using environmental support

Abstract

The invention relates to the related technical field of intelligent control of two-wheeled vehicles, in particular to a method and system for controlling reversing balance of a two-wheeled vehicle by utilizing environmental support. The method comprises the following steps: acquiring environment data of 10-50 meters behind through a collaborative sensing network (V2X module + millimeter wave radar + fisheye camera), and identifying an obstacle and a supporting structure through multi-source fusion; a Z-shaped track is planned to guide the vehicle to avoid obstacles and make contact with a supporting structure to assist balance; based on vehicle body posture data, reverse torque is output through a PD controller to actively maintain balance, and multi-mode warning (handlebar vibration outputs a corresponding mode according to the position of an obstacle, and a tail LED projects a dynamic light band) is matched; a three-level protection mechanism is further arranged, and torque is dynamically adjusted or braking is triggered according to the obstacle distance. The system correspondingly comprises a sensing module, a planning module, a control module and the like. The limitation of a traditional scheme is broken through, and the reversing safety and convenience of the two-wheeled vehicle are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control of two-wheeled vehicles, and particularly relates to a method and system for two-wheeled vehicle reverse balance control using environmental support. BACKGROUND

[0002] Two-wheeled vehicles (including electric bicycles, electric motorcycles, etc.) are often operated in reverse in narrow spaces (such as garages, elevator entrances, and restaurant alleys) due to their flexibility. Unlike traditional four-wheeled vehicles, two-wheeled vehicles face significant balance challenges when reversing at low speeds. Users often need to use one or both feet to maintain balance, which is not only inconvenient to operate, but also extremely easy to lose balance and tip over when holding heavy objects or on wet roads.

[0003] There are some two-wheeled vehicle self-balancing solutions in the prior art, but they are mainly aimed at the forward direction of the vehicle and have limited effect in reverse scenarios. These solutions usually only rely on the vehicle's own gyroscope and motor control to maintain balance, and fail to fully utilize the surrounding environmental features. When encountering sudden obstacles, the system often uses simple deceleration or braking strategies, which may instead disrupt the balance of the vehicle body and increase the risk of falling. SUMMARY

[0004] Therefore, the purpose of the present application is to overcome the shortcomings of the prior art and provide a method and system for two-wheeled vehicle reverse balance control using environmental support to solve the problems of balance difficulty, collision, and tipping over when two-wheeled vehicles reverse in narrow spaces.

[0005] In a first aspect, the present application provides a method for two-wheeled vehicle reverse balance control using environmental support, comprising:

[0006] obtaining surrounding environment data;

[0007] identifying support structures that can be used to stabilize the vehicle body during the reversing process based on the environment data;

[0008] planning a reverse trajectory that guides the two-wheeled vehicle to come into contact with the support structure during the reversing process;

[0009] based on real-time vehicle body posture data and the reverse trajectory, controlling the drive unit to output torque to actively maintain the balance of the vehicle body while guiding the two-wheeled vehicle to move along the reverse trajectory.

[0010] Optionally, identifying support structures comprises:

[0011] analyzing the stability, surface flatness, and height characteristics of surrounding objects based on the environment data, and identifying walls, columns, or guardrails that meet the preset conditions as support structures.

[0012] Optionally, planning a reverse trajectory comprises:

[0013] The Z-shaped trajectory is planned to guide the two-wheeled vehicle to avoid the obstacle while actively approaching the support structure so as to facilitate the user to temporarily support by means of the support structure.

[0014] Optionally, the method further comprises:

[0015] When the distance between the two-wheeled vehicle and the obstacle is less than a first preset safety threshold, the torque of the two-wheeled vehicle is controlled to be not greater than a preset torque;

[0016] When the distance between the two-wheeled vehicle and the obstacle is less than a second preset safety threshold, the two-wheeled vehicle is controlled to enable a reverse balance function;

[0017] When the distance between the two-wheeled vehicle and the obstacle is less than a third preset safety threshold, the two-wheeled vehicle is controlled to cut off power and perform a tactile warning;

[0018] The first preset safety threshold is greater than the second preset safety threshold, and the second preset safety threshold is greater than the third preset safety threshold.

[0019] Optionally, the method further comprises:

[0020] Based on the surrounding environment data, the obstacle is identified;

[0021] Based on the obstacle, a warning is performed and obstacle information is indicated;

[0022] The obstacle includes a vehicle, a pedestrian or an object;

[0023] The obstacle information is a position of the obstacle.

[0024] Optionally, the warning comprises:

[0025] The position information of the obstacle is converted into a phase difference and / or an amplitude difference of left and right handlebar vibration motors;

[0026] The left obstacle corresponds to right handlebar vibration or vibration intensity greater than the left, and the right obstacle corresponds to left handlebar vibration or vibration intensity greater than the right.

[0027] Optionally, the surrounding environment data is obtained by at least one of the following ways:

[0028] The environment data is obtained by a vehicle-mounted environment perception sensor;

[0029] The environment information is received from an external networked device by a V2X communication module.

[0030] In a second aspect, the present application provides a system for two-wheeled vehicle reverse balance control using environmental support, comprising:

[0031] a perception module configured to obtain surrounding environment data;

[0032] an identification module configured to identify, based on the environment data, a support structure available for stabilizing the vehicle body during reversing;

[0033] a planning module configured to plan a reversing trajectory guiding the two-wheeled vehicle to contact the support structure during reversing;

[0034] a control module configured to control the driving unit to output torque based on real-time vehicle body posture data and the reversing trajectory, so as to actively maintain balance of the vehicle body while guiding the two-wheeled vehicle to move along the reversing trajectory.

[0035] Optionally, the system further comprises a haptic feedback device.

[0036] The control module is further configured to control the haptic feedback device to generate a haptic warning signal with directional orientation according to the position and distance information of the identified obstacle.

[0037] Optionally, the system further comprises a V2X communication module configured to receive environment information sent by an external networked device.

[0038] In the scheme provided in the present application, surrounding environment data is obtained; based on the environment data, a support structure available for stabilizing the vehicle body during reversing is identified; a reversing trajectory guiding the two-wheeled vehicle to contact the support structure during reversing is planned; and the driving unit is controlled to output torque based on real-time vehicle body posture data and the reversing trajectory, so as to actively maintain balance of the vehicle body while guiding the two-wheeled vehicle to move along the reversing trajectory. By identifying environmental supports such as walls and columns and actively planning a contact trajectory, the environmental structure is converted into a natural "balance aid tool", greatly reducing the requirement for user balance skills. BRIEF DESCRIPTION OF DRAWINGS

[0039] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description of embodiments of the present application, taken in conjunction with the accompanying drawings. The drawings provided in the present application are used to provide further understanding of the embodiments of the present application, and constitute a part of the specification, and are used to explain the present application together with the embodiments of the present application, but do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0040] Figure 1 is a flowchart of a method for two-wheeled vehicle reversing balance control using environmental support provided by an embodiment of the present application.

[0041] Figure 2is a structural schematic diagram of a method and device for two-wheeled vehicle reverse balance control using environmental support provided by an embodiment of the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0043] The present application provides a revolutionary two-wheeled vehicle reverse balance control solution. Its core idea can be summarized as: "environment is a resource, intelligence to cooperate". It completely jumps out of the framework of the traditional four-wheeled vehicle reverse solution, and surpasses the two-wheeled vehicle solution that only relies on the balance of the vehicle itself, and creatively uses the surrounding environment to realize safe and convenient reversing.

[0044] Specifically, the present application converts two-wheeled vehicle reversing from a "physical work" that requires superb skills to a "easy thing" assisted by system intelligence through the innovation of three levels of perception, planning and control.

[0045] The system not only identifies obstacles, but also actively identifies stable environmental structures such as walls and columns, and regards them as "natural handrails" or "support points" that can be borrowed. Through V2X vehicle networking technology to obtain over-the-horizon information, early warning of curves and blind area oncoming vehicles is realized, breaking through the physical limitations of vehicle-mounted sensors. A warning system based on orientation tactile coding is created. Through the phase difference and intensity difference of left and right handlebar vibration motors, the user is intuitively told "which side has obstacles, how close", the user can perceive the environment without dividing visual attention, and realizes "hand-in-hand" guidance. The system output torque is not only used to drive the vehicle, but also used to actively maintain the balance of the vehicle body, so that the user can maintain stability even when reversing at low speed. Plan a zigzag reverse trajectory, actively guide the vehicle to complete the reversing in stages with support, divide the whole into parts, and reduce the difficulty.

[0046] The scheme provided by the application solves the inherent balance problem of the two-wheeled vehicle when reversing at low speed through the double strategies of "active balance control" and "environment support by taking advantage of the environment". Through "multi-level safety mechanism" and "cooperative perception", the risk of collision and falling caused by blind area and panic operation is greatly reduced. Through "tactile orientation warning", the problem of limited visual attention of the user when reversing and the low efficiency of traditional sound and light warning is solved. In this way, when the two-wheeled vehicle is reversing, the system assists in balancing and the user does not need to exert effort to support the ground. The vehicle can "see" the blind area and "actively protect" the user when in danger. The rear environment can be perceived through handlebar vibration, as if having "tactile vision". The vehicle can plan the optimal route and guide the user to complete the reversing by means of the wall surface and the like.

[0047] After introducing the basic principle of the application, various non-limiting embodiments of the application will be specifically introduced with reference to the accompanying drawings.

[0048] Figure 1 is a flowchart of a method for two-wheeled vehicle reversing balance control using environment support provided by an embodiment of the application. As shown in the figure, the method comprises the following contents. Figure 1

[0049] S101: Obtain surrounding environment data.

[0050] This step obtains environment data through the dual mode of "self-vehicle perception + network cooperation", breaks through the limitation of traditional single sensor detection, ensures the accuracy of data, and the specific implementation mode comprises:

[0051] Self-vehicle sensor perception: the two-wheeled vehicle is equipped with two types of environment perception sensors, millimeter wave radar and fisheye camera:

[0052] The millimeter wave radar collects data such as the distance, horizontal azimuth angle and vertical elevation angle of the rear obstacle at a sampling frequency of not less than 50Hz, and can effectively identify obstacles in adverse environments such as rain, fog and curves; the fisheye camera collects the rear environment image and outputs the pixel coordinates of the obstacle, and the focal length is obtained through a checkerboard calibration board, with a calibration error controlled within 1 pixel to ensure the accuracy of subsequent coordinate conversion.

[0053] V2X network cooperative perception: the two-wheeled vehicle is equipped with a low-power V2X communication module (such as LTE-V2X or NR-V2X), which can receive environment information (including the position coordinates and speed vector of the connected device) sent by other connected devices (such as surrounding vehicles and roadside units) within a range of 10-50 meters in real time, and make up for the line-of-sight blind area of the self-vehicle sensor (such as the incoming vehicle that cannot be directly detected at the corner of the garage).

[0054] ​The multi-source data is pre-processed: the obstacle data collected by the millimeter wave radar is converted into three-dimensional space coordinates; the pixel coordinates output by the fisheye camera are corrected for distortion (to correct the coordinate deviation caused by lens optical distortion), and then the corrected image coordinates are converted into real-world coordinates. Finally, the two types of coordinate data after preprocessing are fused by a specific algorithm to filter data errors and obtain the final position information of the obstacle, ensuring that the position error is not more than 0.1 meters.

[0055] S102: Identify a support structure that can be used to stabilize the vehicle body during reversing based on the environmental data.

[0056] This step screens fixed structures in the surrounding environment that can assist in balancing the vehicle body, converting "environmental obstacles" into "balance resources". The specific identification logic is as follows:

[0057] Screening feature definition: the support structure must meet the three core features of "stability-flatness-contactability":

[0058] In terms of stability, it needs to be a fixed structure such as a wall, a column, a guardrail, etc., and not a moving target such as a pedestrian or a vehicle. Cross-validation is performed through V2X data and image recognition results (e.g., if the object's position changes by no more than 0.05 meters in three consecutive images, it is determined to be a fixed structure);

[0059] In terms of surface flatness, the object's contact surface needs to be close to the vertical ground (the contact surface normal angle is within the range of 80°-100°), which is calculated based on the point cloud data of the millimeter wave radar to ensure that the scooter does not slip when in contact;

[0060] In terms of contactability, the initial distance between the object and the scooter needs to be within the range of 0.5-2.0 meters, and there should be no other obstacles blocking the contact path.

[0061] Support point scoring and determination: a support point scoring mechanism is introduced to quantify the identification results. The score takes into account factors such as object distance, contact surface normal angle, and surface roughness, with priority given to contact safety. When the score reaches 0.8 or above, the object is determined to be a support structure that can be used for reversing balance, and its position coordinates, contact surface range, and other key parameters are recorded to provide a basis for subsequent trajectory planning.

[0062] S103: Plan a reversing trajectory that guides the scooter to come into contact with the support structure during reversing.

[0063] This step ensures that the scooter can establish effective contact with the support structure to assist in balancing while avoiding obstacles through "obstacle avoidance-contact" trajectory design. The specific trajectory planning logic is as follows:

[0064] Trajectory type and parameter setting: Z-shaped trajectory is adopted to realize the action of "first moving away from the obstacle and then approaching the support structure" through lateral swinging to avoid excessive tilting of the vehicle body during reversing. The trajectory parameters are adapted to the scene:

[0065] The reversing speed is set to be no more than 1 m / s (low speed can reduce the difficulty of balance control and meet the demand of reversing in narrow space);

[0066] The trajectory amplitude is positively correlated with the width of the current scene (e.g. when the garage width is 2.0 meters, the amplitude is set to 0.6 meters) to ensure that the maximum lateral swing of the trajectory does not exceed the boundary of the lane and avoid collision with the objects on both sides;

[0067] The swing frequency is set to 0.2 Hz (one lateral swing cycle is completed every 5 seconds) to balance the length of vehicle body tilting and the user's operation burden.

[0068] Trajectory equation and key path points: the initial position of the two-wheeled vehicle when starting reversing is taken as the origin to establish a rectangular coordinate system (x-axis is the reversing direction and y-axis is the lateral horizontal direction), the parameter equation of the Z-shaped trajectory is constructed based on the above parameters to ensure that the vehicle body realizes the periodic action of "moving away from the obstacle-approaching the support structure" along the trajectory.

[0069] At the same time, two core path points are set in the trajectory: obstacle avoidance path point (when the vehicle body reaches the maximum lateral swing away from the obstacle) and contact path point (when the vehicle body reaches the maximum lateral swing approaching the support structure) to ensure the safety and balance assistance effect of the reversing process, such as the safety distance between the vehicle body and the obstacle at the obstacle avoidance path point is not less than 0.8 meters and the vehicle body can stably contact the support structure at the contact path point.

[0070] S104: Based on the real-time vehicle body posture data and the reversing trajectory, the driving unit output torque is controlled to actively maintain the balance of the vehicle body while guiding the two-wheeled vehicle to move along the reversing trajectory.

[0071] This step realizes the dual goals of "trajectory tracking + vehicle body balance" through torque control to ensure that the two-wheeled vehicle does not tip over while moving along the planned trajectory. The specific control logic is as follows:

[0072] Vehicle body posture data acquisition: through the gyroscope (vehicle body posture sensor) carried by the two-wheeled vehicle, the vehicle body inclination angle (reflecting the degree of vehicle body tilting, the value range is controlled within -10°-10°, and if it exceeds the range, it is determined as a tilting risk) and inclination rate (reflecting the vehicle body tilting speed) are collected in real time. The sampling frequency is 100 Hz to ensure the real-time nature of the posture data and provide a basis for balance control.

[0073] Torque calculation and output: the PD controller (proportional-derivative controller) is used to calculate the compensation torque of the rear motor (drive unit) to achieve body balance control. The parameters of the PD controller are optimized and set, the proportional parameter is used to quickly respond to the body tilt angle deviation and reduce static error, and the differential parameter is used to suppress the tilt angle rate and avoid excessive body swing.

[0074] The calculated compensation torque is a reverse torque (opposite to the forward direction torque, driving the vehicle in reverse), while setting a maximum auxiliary torque limit (not more than 5Nm) to avoid sudden acceleration or tilting of the vehicle body caused by excessive torque; when the two-wheeled vehicle contacts with the supporting structure, the torque is reduced to below 2Nm to prevent excessive dependence on the supporting structure from causing a collision.

[0075] Balance control closed loop and safety constraints: real-time comparison of the actual trajectory of the two-wheeled vehicle with the planned trajectory, and the actual body tilt angle with the target tilt angle (which is dynamically adjusted according to the reverse speed and distance from the obstacle), if the deviation exceeds the allowed range (such as the actual trajectory deviates from the planned trajectory by more than 0.1 meters, and the actual body tilt angle exceeds the range of -5°-5°), adjust the compensation torque output by the PD controller until the deviation returns to the allowed range, forming a balance control closed loop of "perception-calculation-execution-feedback".

[0076] In addition, a safety constraint mechanism is provided: when the two-wheeled vehicle is less than 0.3 meters away from other obstacles outside the supporting structure (the first preset safety threshold), or the two-wheeled vehicle approaches the supporting structure in a way that does not meet the preset safety conditions (such as the body tilt angle exceeds 10° when contacting), the active balance control is immediately stopped, and the emergency brake (rear motor output brake torque) is triggered to prevent collision or overturning.

[0077] In some embodiments, the supporting structure is identified, including:

[0078] According to the analysis of the stability, surface flatness and height characteristics of the surrounding objects based on the environmental data, the wall, column or guardrail that meets the preset conditions is identified as the supporting structure.

[0079] This claim further specifies the specific determination criteria and preset conditions of "stability, surface flatness, height characteristics" to ensure the accuracy and reliability of the supporting structure identification, and to avoid misjudging moving targets or inappropriate structures as supporting resources. The specific description is as follows:

[0080] 1. Surrounding object feature analysis dimension

[0081] (1) Stability determination

[0082] Stability is the core prerequisite for supporting structures, which needs to ensure that the identified object is a fixed structure with no risk of movement. The specific determination method is as follows:

[0083] Cross-validation of data sources: Combining continuous frame data from millimeter-wave radar with V2X communication information, if the positional offset of an object in 3-5 consecutive frames of data (time interval 0.1-0.2 seconds) is ≤0.05 meters, and the V2X module does not detect the velocity vector of the object (or the velocity is ≤0.1m / s), it is determined to be a "stable fixed structure"; if the positional offset of the object is detected to be >0.05 meters (such as pedestrians or moving vehicles), it is directly excluded as a supporting structure.

[0084] Scene experience rule adaptation: In scenarios such as garages and alleyways in residential areas, prioritize including walls, concrete columns, and fixed guardrails in the stability judgment scope; in temporary roads or densely populated areas (such as food streets), only structures with their bottoms fixed to the ground and a height of ≥1.2 meters (such as fixed road bollards) are included in the judgment to avoid misjudging temporarily placed items (such as cardboard boxes and plastic cones).

[0085] (2) Surface flatness judgment

[0086] The surface flatness must meet the requirement of "not easily slipping when two wheels are in contact and providing stable support". Specific judgment criteria:

[0087] Contact surface morphology requirements: Through image texture analysis using a fisheye camera, if the surface unevenness difference is ≤5mm (such as a flat wall or smooth column), it is judged as "flat surface"; if there are obvious protrusions (such as wall pipes or column decoration protrusions) or depressions (such as gaps in a damaged wall), and the unevenness difference is >5mm, it is excluded as a supporting structure because it is easy to cause the vehicle body to tilt or slip.

[0088] Normal angle auxiliary verification: Combine the point cloud data of millimeter-wave radar to fit the normal direction of the contact surface. If the angle between the normal and the ground (surface normal angle) is in the range of 80°-100° (close to being perpendicular to the ground), the surface flatness is further confirmed to meet the requirements. If the normal angle is <80° (such as a sloping surface) or >100° (such as a concave arc surface), it does not meet the preset conditions because it cannot provide a support force perpendicular to the vehicle body.

[0089] (3) Height feature determination

[0090] The height feature must match the body structure of a two-wheeled vehicle to ensure that the supporting structure can make effective contact with the side of the vehicle body (or handlebars). Specific judgment criteria:

[0091] Height range setting: The height of the support structure should cover the "balance critical area" of the two-wheeled vehicle, which is the range from 0.6 to 1.2 meters above the ground (corresponding to the middle of the two-wheeled vehicle body to the handlebar height). If the structure height is less than 0.6 meters (such as a low roadside), it cannot provide support to the middle of the vehicle body. If the height is greater than 1.8 meters (such as a high wall), only the top can be contacted, which is easy to cause the vehicle body to fall, and does not meet the preset conditions.

[0092] Height data acquisition: Calculate the height of the object through the vertical elevation angle data of the millimeter wave radar (combined with the object distance and elevation angle range), or through the image proportion conversion of the fisheye camera (with the height of the two-wheeled vehicle body as the reference benchmark), to ensure that the height determination error is less than or equal to 0.1 meters.

[0093] 2. Preset conditions and identification result output

[0094] When an object meets the following preset conditions at the same time, it is determined as a "support structure for balancing when reversing a vehicle":

[0095] Stability: The position offset of consecutive frames is less than or equal to 0.05 meters, and it is a fixed structure (not a moving target);

[0096] Surface flatness: The surface concave-convex fluctuation difference is less than or equal to 5mm, and the normal angle of the contact surface is 80°-100°;

[0097] Height characteristics: The height covers the range of 0.6-1.2 meters, and there is no obstruction to block the contact path between the vehicle body and the structure.

[0098] After the identification is completed, the system automatically records the position coordinates of the support structure, the contact surface range (width, height), and marks it as a "green support point" on the simple map of the vehicle-mounted instrument, providing a clear target for subsequent trajectory planning.

[0099] In some embodiments, planning a reversing trajectory includes: planning a zigzag trajectory to guide the two-wheeled vehicle to actively approach the support structure while avoiding obstacles, so as to facilitate the user to temporarily support the support structure.

[0100] The essence of the zigzag trajectory is to achieve two key goals during the reversing process through "lateral periodic oscillation":

[0101] Obstacle avoidance goal: By oscillating away from the obstacle, the vehicle body and the obstacle always maintain a safe distance (≥0.8 meters), avoiding collision due to blind area during reversing;

[0102] Support contact target: By swinging towards the support structure, the side of the vehicle body (such as the handlebars, body guards) can make slight contact with the support structure, and the user can maintain balance with the help of the reaction force of the structure - without the user's feet supporting the ground, only through the temporary contact of the hands or vehicle body with the support structure, the tendency of the vehicle body to tilt can be offset.

[0103] 2. Adaptability design of trajectory planning

[0104] (1) Swing direction related to the position of obstacles / support structures

[0105] The swing direction of the trajectory needs to be dynamically adjusted according to the relative position of the obstacles and support structures:

[0106] If the obstacle is on the left side of the two-wheeled vehicle (such as the left pillar of the garage), and the support structure is on the right side (such as the right wall of the garage), the swing direction of the Z-shaped trajectory is "first swing to the right (away from the left obstacle) → then swing to the left (close to the right support structure)";

[0107] If the positions of the obstacle and the support structure are reversed (obstacle on the right, support structure on the left), the swing direction is reversed to ensure that the "away" and "close" actions of each swing accurately correspond to the target.

[0108] (2) Matching of trajectory parameters and support contact requirements

[0109] To ensure that the user can easily achieve temporary support with the help of the support structure, the trajectory parameters need to meet the requirements of "moderate contact force and low operation difficulty":

[0110] Swing amplitude control: The maximum lateral swing amplitude (amplitude) of the trajectory is set to 0.3-0.8 meters, which is adjusted according to the initial distance between the support structure and the vehicle body - the closer the initial distance, the smaller the amplitude (such as an initial distance of 0.5 meters, an amplitude of 0.3 meters), to avoid rapid tilting of the vehicle body due to excessive swing amplitude; the farther the initial distance, the amplitude is appropriately increased (such as an initial distance of 1.2 meters, an amplitude of 0.6 meters), to ensure that the support structure can be quickly approached.

[0111] Contact timing design: Within each swing cycle (5 seconds) of the trajectory, a "stable contact period" of 1-2 seconds is set - when the vehicle body swings to the limit position close to the support structure, the lateral speed of the trajectory is temporarily reduced (≤0.1 m / s), giving the user 1-2 seconds to adjust balance with the help of the contact between the vehicle body and the support structure, to avoid insufficient support due to too short contact time.

[0112] 3. User assistance prompts

[0113] To help users understand the trajectory intent and cooperate with the support contact action, the system provides auxiliary prompts through the vehicle-mounted instrument and voice:

[0114] When the trajectory is about to swing to the support structure, the instrument displays "about to approach the right support point, please prepare to touch the balance lightly", and the voice broadcast "right support, auxiliary balance";

[0115] When the vehicle body is in contact with the support structure, the instrument displays "has contacted the support point, can adjust the vehicle body with the help of support", to avoid user misoperation due to not noticing the contact.

[0116] In some embodiments, when the distance between the two-wheeled vehicle and the obstacle is less than a first preset safety threshold, the torque of the two-wheeled vehicle is controlled to be not greater than a preset torque; when the distance between the two-wheeled vehicle and the obstacle is less than a second preset safety threshold, the two-wheeled vehicle is controlled to enable a reverse balance function; when the distance between the two-wheeled vehicle and the obstacle is less than a third preset safety threshold, the two-wheeled vehicle is controlled to cut off power and perform tactile warning; wherein the first preset safety threshold is greater than the second preset safety threshold; and the second preset safety threshold is greater than the third preset safety threshold.

[0117] According to the risk level of the two-wheeled vehicle reversing scene, three safety thresholds from large to small are set, covering "safe-risk-emergency" three states:

[0118] The first preset safety threshold is set to 0.8-1.2 meters (preferably 1.0 meter), corresponding to a "low-risk warning state" - at this time, the two-wheeled vehicle is far away from the obstacle, and there is no need for emergency braking, but the torque needs to be limited to prevent the speed from being too fast;

[0119] The second preset safety threshold is set to 0.4-0.6 meters (preferably 0.5 meters), corresponding to a "medium-risk intervention state" - at this time, the distance to the obstacle is close, and the reverse balance function needs to be enabled to enhance the stability of the vehicle body to avoid tipping due to emergency steering;

[0120] The third preset safety threshold is set to 0.2-0.3 meters (preferably 0.3 meters), corresponding to a "high-risk emergency state" - at this time, the distance to the obstacle is extremely close, and power needs to be immediately cut off and warned to prevent collision.

[0121] The size relationship of the three thresholds strictly satisfies: the first preset safety threshold > the second preset safety threshold > the third preset safety threshold, ensuring the progressiveness and rationality of the control strategy.

[0122] 2. Control action details corresponding to different thresholds

[0123] (1) First preset safety threshold triggering (distance <1.0m, ≥0.5m): torque limitation control

[0124] Control target: limit the output torque of the driving unit, reduce the reversing speed, and reserve reaction time for subsequent obstacle avoidance or braking;

[0125] Specific action: Limit the maximum output torque of the drive unit to "preset torque" (set to 2-3 Nm, about 50% of normal reverse torque), at this time the two-wheeled vehicle reverse speed decreases from ≤1 m / s to ≤0.5 m / s; At the same time, the vehicle-mounted instrument displays "close to the obstacle, the speed has been limited", reminding the user to slow down.

[0126] (2) Second preset safety threshold trigger (distance <0.5 meters, ≥0.3 meters): reverse balance function enabled

[0127] Reverse balance function definition: By adjusting the torque direction and size of the rear wheel motor, actively offset the tilting trend of the vehicle body to the obstacle side, enhance the balance stability;

[0128] Specific action: If the system detects that the vehicle body tilts to the obstacle side (such as the obstacle is on the left side, the vehicle body inclination angle deviates to the left >3°), the drive unit outputs reverse torque (opposite to the normal reverse torque direction, size 1-2 Nm), assisting the vehicle body to return to normal; At the same time, the handlebar vibration motor emits "one-sided low-frequency vibration" (such as the obstacle is on the left side, the right handlebar vibrates, prompting the user to adjust the direction to the right), guiding the user to actively avoid obstacles.

[0129] (3) Third preset safety threshold trigger (distance <0.3 meters): power cut-off and tactile alarm control target: immediately stop the reverse action, and remind the user to handle urgently through strong alarm;

[0130] Specific action: ① Cut off the power output of the drive unit, prohibit any reverse torque output; ② Start the handlebar double-sided vibration motor to emit an emergency alarm with 40Hz high-frequency vibration (human sensitive frequency), and the vehicle-mounted buzzer emits a 1000Hz high-frequency beep; ③ If the two-wheeled vehicle is equipped with a rear LED projection, immediately switch the projection light strip from "yellow" to "red flashing", reminding the rear pedestrians or vehicles to avoid.

[0131] 3. Dynamic switching and resetting of threshold trigger

[0132] Switching logic: The system calculates the obstacle distance in real time through millimeter wave radar and fusion data, when the distance decreases from "≥1.0 meters" to "<1.0 meters", the first threshold control is triggered; When the distance continues to decrease to "<0.5 meters", automatically switch to the second threshold control; When the distance decreases to "<0.3 meters", switch to the third threshold control, without the need for user manual intervention;

[0133] Resetting logic: If the user operates the two-wheeled vehicle away from the obstacle (the distance rises to ≥1.0 meters), the system releases the control actions corresponding to the third, second, and first thresholds in turn, restores the normal reverse torque and speed, and ensures the operation flexibility.

[0134] In some embodiments, the method further comprises: based on the surrounding environment data; identifying an obstacle; based on the obstacle, issuing a warning and indicating obstacle information; the obstacle includes a vehicle, a pedestrian or an object; the obstacle information is the position of the obstacle.

[0135] Based on the cooperative perception network (self-sensor + V2X), the obstacle identification range covers the area of 10-50 meters behind and laterally behind the two-wheeled vehicle, and focuses on the target area that may affect the safety of reversing, i.e. the objects within 1.5 meters on both sides of the reversing track and within 3 meters behind.

[0136] According to the moving characteristics and risk level of the object, the obstacles are divided into three categories, and different types correspond to different warning priorities.

[0137] The first type: moving obstacles, including pedestrians, non-motor vehicles (bicycles, electric vehicles), motor vehicles (cars, trucks), such obstacles have dynamic characteristics, the risk level is the highest, and the warning priority is the first;

[0138] The second type: fixed obstacles, including road edges, stone piers, trees, walls (non-supporting structures), such obstacles have fixed positions, but are easily ignored due to visual blind area, and the warning priority is the second;

[0139] The third type: low obstacles, including ground depressions, protruding manhole covers, and scattered debris, such obstacles have a height of <0.3 meters, are easily scratched by the chassis of the two-wheeled vehicle, and the warning priority is the third.

[0140] The core information of the obstacle, i.e. the position, is obtained through multi-source data fusion, and the specific implementation manner is as follows:

[0141] Self-sensor analysis: millimeter wave radar directly determines the position of the obstacle through horizontal azimuth angle (such as azimuth angle -30° representing left rear 30°, and azimuth angle 60° representing right rear 60°); fish-eye camera converts the position by combining the pixel position of the obstacle in the image with the calibration parameters (such as the pixel being located on the left side of the image, corresponding to the left rear obstacle);

[0142] V2X information supplement: the obstacle information sent by the external networking device (such as a roadside unit) contains a “relative position” field (such as “the target is located 20 meters right rear of your vehicle”), and the system directly analyzes the field to supplement the position data;

[0143] Uniform representation of position: all position information from different sources is uniformly converted into the format of “taking the center axis of the two-wheeled vehicle as the reference, left rear as negative angle (-180°-0°), and right rear as positive angle (0°-180°)”, to ensure the consistency of the information.

[0144] According to the type and position of the obstacle, a "layered warning" strategy is adopted to ensure that the user can quickly perceive while maintaining balance:

[0145] The first type of moving obstacle: adopt "voice + tactile" dual warning, such as a pedestrian approaching from the right rear, voice broadcast "pedestrian behind the right, please be careful to avoid", and left handlebar vibration (prompting the user to turn left to avoid obstacles);

[0146] The second type of fixed obstacle: adopt "tactile + instrument display" warning, such as a stone pillar to the left rear, right handlebar vibration, and instrument display "fixed obstacle to the left rear, keep distance";

[0147] The third type of low obstacle: adopt "instrument display + buzzer" warning, such as a depression in the rear ground, instrument display "depression in the rear ground, slow down to pass", and buzzer at 500Hz low frequency.

[0148] The intensity of the warning increases as the distance to the obstacle decreases (such as the distance to the moving obstacle decreases from 5 meters to 2 meters, the voice volume increases from 60 decibels to 80 decibels, and the vibration frequency increases from 20Hz to 40Hz), ensuring that the user can perceive the risk level.

[0149] In some embodiments, the warning includes: converting the position information of the obstacle into the phase difference and / or amplitude difference of the left and right handlebar vibration motors; wherein the left obstacle corresponds to the right handlebar vibration or the vibration intensity is greater than the left, and the right obstacle corresponds to the left handlebar vibration or the vibration intensity is greater than the right.

[0150] The implementation of the vibration mode depends on the two-wheel vehicle handlebar built-in double independent linear vibration motors (one for each left and right handlebar), with the following hardware parameters: vibration frequency range: 20Hz-50Hz (covering the human touch sensitive frequency band, avoiding too low frequency leading to unclear perception and too high frequency causing hand discomfort); amplitude adjustment range: 0.1g-0.3g (g is the acceleration of gravity, the amplitude directly corresponds to the vibration intensity, 0.1g is "weak vibration", 0.2g is "moderate vibration", and 0.3g is "strong vibration"); phase difference control accuracy: ≤10ms (i.e. the left and right motor start time difference can be accurately controlled to 0.01 seconds, ensuring that the user can clearly perceive the "single side leading" position difference); response delay: ≤50ms (the time difference from system recognition of obstacle position to motor start vibration, meeting the real-time warning needs in reverse scenarios).

[0151] The detailed definition of the vibration mode and perception correspondence is as follows:

[0152] 2.3 Vibration mode and perception correspondence

[0153]

[0154] The scheme provided by the application also includes: controlling a projection device arranged at the tail of the two-wheeled vehicle to project a dynamic light strip on the ground as a moving safety warning area.

[0155] The LED projection module is installed at the tail of the two-wheeled vehicle, and the hardware needs to meet the following requirements: RGB three-color LED lamp beads (luminance ≥ 500 lm, supporting red / yellow / green switching) are used, the horizontal projection angle is 60°±5°, the vertical projection distance is 0.5-3 m, the light strip width is 15-20 cm, the protection level is IP67 to adapt to outdoor environment, the response delay is ≤100 ms, and real-time warning is ensured. The projection direction is consistent with the central axis of the vehicle body, and the vertical included angle is 15°±3°, so as to avoid light strip deviation or too close / too far.

[0156] The dynamic light strip matches the obstacle distance in terms of "color+length+state" to form a moving safety warning area, and the specific rules are as follows:

[0157] Green light strip: obstacle distance > 1.5 m (low risk), 0.5-3 m complete light strip is projected, and the rear safety is prompted;

[0158] Yellow light strip: obstacle distance 0.8-1.5 m (medium risk), light strip is shortened to 0.5-2 m, and the rear needs to be avoided;

[0159] Red light strip: obstacle distance < 0.8 m (high risk), light strip is shortened to 0.5-1 m and 2 Hz flickering, and the rear is prohibited to enter the light strip area.

[0160] The light strip moves synchronously with the reversing action, dynamically adjusts the length and color with the change of the obstacle distance, and simultaneously links the handlebar vibration feedback (such as red light strip synchronously triggering double-sided high-frequency strong vibration), to form "visual+touch" dual warning.

[0161] The scheme provided by the embodiment is suitable for scenes such as catering streets and narrow garages. When a delivery rider reverses, the light strip can prompt pedestrians behind to avoid; when reversing in a garage, the light strip can assist the user in judging the distance from the wall and column, reducing the risk of collision, and solving the problem that the traditional two-wheeled vehicle can only rely on the buzzer "to fail to warn others and be difficult to distinguish the risk level".

[0162] In summary, the scheme provided in the application empowers group perception through V2X communication: through a low-power V2X module (such as LTE-V2X or NR-V2X), a vehicle in reverse can obtain data of other networked devices within a range of 10-50 meters behind it in real time. For example, when a vehicle with map navigation is coming from the opposite direction, the system receives its position and speed vector in advance, displays a dynamic icon of "oncoming vehicle" on the instrument UI, and issues a beeping warning. This scheme breaks through the range limitation of self-sensor and is particularly suitable for scenes where traditional sensors fail, such as curves and rainy and foggy weather. Haptic feedback and gravity center guidance are used: a linear vibration motor is built into the handlebar to indicate the position of the obstacle through vibration frequency and direction. Left-side vibration indicates an obstacle behind the right (prompting a left turn to avoid), and bilateral high-frequency vibration indicates emergency braking. The biggest difference from the four-wheel scheme is the linkage with gravity center control: when the system detects that the user is leaning to the side of the obstacle, it automatically fine-tunes the rear wheel torque to assist the vehicle body to return to normal, avoiding the risk of falling due to panic. This module transmits obstacle position information through haptic coding, and actively compensates for gravity center deviation using micro-torque control, achieving coordinated control of warning and balance. The following is the detailed working principle and formula expression: in the scheme provided in the application, phase difference coding is used to realize "which side vibrates first = which side has an obstacle"; intensity modulation is used to transmit distance information (the closer, the stronger). Intelligent gravity center adjustment based on obstacle position. By identifying environmental supports such as walls and pillars, and actively planning contact trajectories, the environmental structure is converted into a natural "balance aid tool", greatly reducing the requirement for user balance skills.

[0163] The system embodiment of the application can be used to execute the method embodiment of the application. For details not disclosed in the system embodiment of the application, please refer to the method embodiment of the application.

[0164] Figure 2 Fig. 1 shows a block diagram of a system for two-wheeled vehicle reverse balance control using environmental support provided by an embodiment of the application. As shown in the figure, Figure 2 The system comprises:

[0165] A perception module 21 is configured to obtain surrounding environment data;

[0166] The perception module comprises a V2X communication module configured to receive environment information sent by external networked devices

[0167] An identification module 22 is configured to identify, based on the environment data, a support structure that can be used to stabilize the vehicle body during the reverse process;

[0168] A planning module 23 is configured to plan a reverse trajectory that guides the two-wheeled vehicle to contact the support structure during the reverse process;

[0169] The control module 24 is configured to control the driving unit to output a torque based on the real-time vehicle body posture data and the reverse trajectory, so as to actively maintain the balance of the vehicle body while guiding the two-wheeled vehicle to move along the reverse trajectory.

[0170] In some embodiments, the system further comprises a haptic feedback device; and the control module is further configured to control the haptic feedback device to generate a haptic warning signal with a directionality according to the orientation and distance information of the identified obstacle.

[0171] The above description has been presented for the purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to forms disclosed herein. Although several example aspects and embodiments have been discussed above, those of ordinary skill in the art will appreciate a variety of modifications, alternatives, permutations, additions, and sub-combinations of the described aspects and embodiments.

Claims

1. A method for reversing balance control of a two-wheeled vehicle using environmental support, characterized in that, include: Acquire surrounding environmental data; Based on the environmental data, support structures that can be used to stabilize the vehicle body during reversing were identified; Plan a reversing trajectory that guides the two-wheeled vehicle to contact the supporting structure during reversing; Based on real-time vehicle posture data and the reversing trajectory, the drive unit outputs torque to actively maintain the vehicle's balance while guiding the two-wheeled vehicle to move along the reversing trajectory.

2. The method for controlling the reversing balance of a two-wheeled vehicle using environmental support according to claim 1, characterized in that, Identify supporting structures, including: Based on the analysis of the environmental data, the stability, surface flatness, and height characteristics of surrounding objects are analyzed, and walls, columns, or railings that meet the preset conditions are identified as supporting structures.

3. The method for controlling the reversing balance of a two-wheeled vehicle using environmental support according to claim 1, characterized in that, Planning the reversing trajectory includes: Plan a zigzag trajectory to guide the two-wheeled vehicle to avoid obstacles while actively approaching the supporting structure so that the user can use the supporting structure for temporary support.

4. The method for controlling the reversing balance of a two-wheeled vehicle using environmental support according to claim 1, characterized in that, The method further includes: When the distance between the two-wheeled vehicle and the obstacle is less than a first preset safety threshold, the torque of the two-wheeled vehicle is controlled to not exceed a preset torque. When the distance between the two-wheeled vehicle and the obstacle is less than the second preset safety threshold, the two-wheeled vehicle is controlled to activate the reverse balancing function. When the distance between the two-wheeled vehicle and the obstacle is less than the third preset safety threshold, the two-wheeled vehicle is controlled to cut off power and issue a tactile alarm. Wherein, the first preset security threshold is greater than the second preset security threshold; the second preset security threshold is greater than the third preset security threshold.

5. The method for controlling the reversing balance of a two-wheeled vehicle using environmental support according to claim 1, characterized in that, The method further includes: Based on the surrounding environment data; and by identifying obstacles; Based on the obstacles, issue warnings and indicate obstacle information; The obstacles include vehicles, pedestrians, or objects; The obstacle information refers to the location of the obstacle.

6. The method for controlling the reversing balance of a two-wheeled vehicle using environmental support according to claim 5, characterized in that, Provide warnings, including: The location information of the obstacle is converted into the phase difference and / or amplitude difference of the vibration motors on the left and right handlebars; Among them, obstacles on the left correspond to vibrations on the right handlebars or vibrations with greater intensity than those on the left, and obstacles on the right correspond to vibrations on the left handlebars or vibrations with greater intensity than those on the right.

7. The method for controlling the reversing balance of a two-wheeled vehicle using environmental support according to claim 1, characterized in that, The acquisition of surrounding environmental data includes at least one of the following methods: Environmental data is acquired through onboard environmental perception sensors; The V2X communication module receives environmental information sent from external networked devices.

8. The method according to claim 1, characterized in that, Also includes: The control unit located at the rear of the two-wheeled vehicle projects a dynamic light strip onto the ground as a moving safety warning zone.

9. A system for reversing balance control of a two-wheeled vehicle using environmental support, characterized in that, The system includes: The sensing module is used to acquire data about the surrounding environment; The sensing module includes a V2X communication module for receiving environmental information sent from external networked devices. The identification module is used to identify, based on the environmental data, supporting structures that can be used to stabilize the vehicle body during reversing; The planning module is used to plan a reversing trajectory that guides the two-wheeled vehicle to contact the supporting structure during the reversing process; The control module is used to control the output torque of the drive unit based on real-time vehicle posture data and the reversing trajectory, so as to actively maintain the vehicle's balance while guiding the two-wheeled vehicle to move along the reversing trajectory.

10. The system according to claim 9, characterized in that, The system also includes a haptic feedback device; The control module is also used to control the tactile feedback device to generate a tactile warning signal with directional orientation based on the location and distance information of the identified obstacles.

Images