An interaction method for a wearable multi-modal tactile feedback system for cyclists

Abstract

The present invention discloses an interaction method for a wearable multimodal tactile feedback system for cyclists, including: obtaining rear moving object data through the lidar, and constructing a proximity danger value function based on the moving object data; obtaining a traffic complexity value through real-time traffic data and weather data analysis, adjusting the weight of the proximity danger value function through the traffic complexity value, and calculating to obtain a proximity danger value; classifying the danger level through the proximity danger value, and obtaining a first multimodal tactile feedback mode according to the danger level; obtaining human physiological data through the human physiological sensor, and adjusting the first multimodal tactile feedback mode through the human physiological data to obtain a second multimodal tactile feedback mode.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-modal haptic feedback, and particularly to an interaction method for a wearable multi-modal haptic feedback system for cyclists. Background Art

[0002] In the modern urban traffic environment, as vulnerable traffic participants, cyclists face many safety challenges on the road. Research shows that the environmental perception ability of cyclists directly affects their cycling safety.

[0003] Traditional environmental perception means mainly rely on visual information, such as road signs, signal lights, and rearview mirrors. These means have problems of untimely information acquisition and limited vision. When cyclists are moving quickly, it is difficult to obtain the changes in the surrounding environment in real time. Therefore, when facing emergencies, cyclists often cannot make timely responses, resulting in potential safety hazards. In addition, in a complex traffic environment, cyclists are faced with too much information, and the overload of visual information may lead to cognitive burden and increase the risk of slow reaction. Cyclists need to quickly identify and respond to various information in a rapidly changing environment, which poses higher requirements for their vision and cognitive abilities.

[0004] Existing haptic feedback systems, such as perception helmets, although have improved the environmental perception of cyclists to a certain extent, most systems are still limited to a single feedback mode, only vibrating to remind the direction or distance of surrounding vehicles through simple vibration, and cannot accurately give corresponding reminders according to the speed or movement trajectory of approaching surrounding vehicles. In addition, in a complex traffic environment, due to the large number of surrounding vehicles, the feedback frequency will increase, but too frequent or strong feedback may cause it difficult for cyclists to distinguish key haptic information, resulting in information interference and inability to judge the actual traffic state.

[0005] Designing an interaction method for a wearable multi-modal haptic feedback system for cyclists to solve the above problems existing in the prior art is the purpose of the research of the present invention. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to propose an interaction method for a wearable multi-modal haptic feedback system for cyclists, which can solve the above problems.

[0007] The present invention provides an interaction method for a wearable multi-modal haptic feedback system for cyclists, based on a wearable multi-modal haptic feedback system for cyclists, including:

[0008] A wearable feedback vest, which is sleeved on the upper body of the cyclist and is used to generate haptic feedback signals on the upper body of the cyclist;

[0009] A laser radar is arranged on the rear side of the wearable feedback vest and connected to the controller to obtain data of moving objects behind the rider;

[0010] A human physiological sensor is disposed on a wearable feedback vest and connected to a controller for collecting human physiological data;

[0011] A tactile feedback element group is distributed on the back side of the wearable feedback vest and connected to the controller for providing multi-modal tactile feedback to the rider;

[0012] The method comprises:

[0013] Acquiring rear-side moving object data through the laser radar, and constructing a proximity risk value function according to the moving object data;

[0014] The traffic complexity value is obtained by analyzing real-time traffic data and weather data, and the weight of the proximity danger value function is adjusted according to the traffic complexity value to calculate the proximity danger value;

[0015] Classifying the danger level by approaching the danger value, and obtaining a first multimodal tactile feedback mode according to the danger level;

[0016] Human physiological data is acquired through the human physiological sensor, and the first multimodal tactile feedback mode is adjusted according to the human physiological data to obtain a second multimodal tactile feedback mode.

[0017] Further, the acquiring rear moving object data by the laser radar and constructing a proximity risk value function according to the moving object data comprises:

[0018] The relative distance of the moving object behind is obtained by the laser radar ,speed , acceleration ,volume ;

[0019] Relative distance of the object moving behind ,speed , acceleration ,volume Constructing a function close to the danger value , the calculation formula is as follows:

[0020] ,

[0021] in, The weight of The weight of The weight of The weight of .

[0022] Furthermore, the traffic complexity value obtained by analyzing real-time traffic data and weather data includes:

[0023] Obtain vehicle density data, road congestion index data, and average vehicle speed data;

[0024] Obtain visibility data, rainfall data, and wind speed data;

[0025] The real-time traffic data and weather data are input into the pre-trained traffic complexity prediction model to obtain the traffic complexity value C.

[0026] Furthermore, the traffic complexity prediction model is trained through the following steps:

[0027] Collect historical real-time traffic data and weather data, mark traffic complexity target values ​​based on the historical real-time traffic data and weather data, and construct training samples;

[0028] The traffic complexity prediction model is constructed through the XGBoost model, and the initial training samples are fitted through the XGBoost model decision tree to initialize the traffic complexity prediction value;

[0029] Based on the loss function, the gradient of each sample is calculated, and a new decision tree is constructed using the gradient and the traffic complexity target value;

[0030] The traffic complexity prediction value of the new decision tree is weightedly summed with the traffic complexity prediction value of the previous round, and the traffic complexity prediction model is iteratively updated.

[0031] Further, adjusting the weight of the approach risk value function by the traffic complexity value includes:

[0032] When the traffic complexity value C is greater than 0.5, increase Weight , Weight , the calculation formula is as follows:

[0033] ,

[0034] in, for The initial weight of for The initial weight of

[0035] When the traffic complexity value C is less than 0.5, increase Weight ,Increase Weight , the calculation formula is as follows:

[0036] ,

[0037] Among them, is the initial weight of is the initial weight of ;

[0038] Normalize and adjust the weight of , the weight of , the weight of , the weight of to obtain the weights of after planning and adjustment , the weight of , the weight of , the weight of .

[0039] Furthermore, the haptic feedback element group includes: a vibration motor group, a piezoelectric element group, and a thermoelectric element group;

[0040] The vibration motor group is respectively arranged at the upper part of the back of the wearable feedback vest and is respectively used for directional haptic feedback of the closest rear moving object;

[0041] The piezoelectric element group is respectively arranged in the middle part of the back of the wearable feedback vest and is respectively used for distance haptic feedback of the closest rear moving object;

[0042] The thermoelectric element group is respectively arranged at the waist of the wearable feedback vest and is respectively used for speed haptic feedback of the closest rear moving object.

[0043] Furthermore, the method for classifying the danger level by approaching the danger value and obtaining the first multi-modal haptic feedback mode according to the danger level includes:

[0044] Those with an approaching danger value exceeding the first threshold are classified as high danger levels, and relative distance feedback of the closest rear moving object is performed through the piezoelectric element group, speed feedback of the closest rear moving object is performed through the thermoelectric element group, and direction feedback of the closest rear moving object is performed through the vibration motor group;

[0045] Those with an approaching danger value lower than the first threshold and exceeding the second threshold are classified as medium danger levels, and relative distance feedback of the closest rear moving object is performed through the piezoelectric element group, and direction feedback of the closest rear moving object is performed through the vibration motor group;

[0046] Those with a proximity to the danger value equal to or lower than the second threshold are classified as a low danger level, and the direction of the closest moving object at the rear is feedback through the vibration motor group.

[0047] Furthermore, the human physiological sensor includes: a heart rate sensor and a respiratory rate sensor;

[0048] The heart rate sensor is disposed at the front chest of the wearable feedback vest for collecting the heart rate data of the cyclist;

[0049] The respiratory rate sensor is disposed at the front chest of the wearable feedback vest for collecting the respiratory rate data of the cyclist.

[0050] Furthermore, obtaining human physiological data through the human physiological sensor and adjusting the first multi-modal tactile feedback mode through the human physiological data to obtain the second multi-modal tactile feedback mode includes:

[0051] Analyzing the current physiological state of the human body through the heart rate data and the respiratory rate data, and the current physiological state of the human body includes: a tense state, a fatigued state, and a relaxed state;

[0052] If the current physiological state of the human body is a tense state, then the direction of the closest moving object at the rear is feedback through the vibration motor group at all danger levels;

[0053] If the current physiological state of the human body is a fatigued state, then the relative distance of the closest moving object at the rear is feedback through the piezoelectric element group, and the direction of the closest moving object at the rear is feedback through the vibration motor group at all danger levels;

[0054] If the current physiological state of the human body is a relaxed state, then the first multi-modal tactile feedback mode is retained.

[0055] Furthermore, analyzing the current physiological state of the human body through the heart rate data and the respiratory rate data includes:

[0056] If the number of heartbeats per minute > 120 and the number of breaths per minute > 20, then the current physiological state of the human body is a tense state;

[0057] If the number of heartbeats per minute < 60 and the number of breaths per minute < 12, then the current physiological state of the human body is a fatigued state;

[0058] If 60 ≤ the number of heartbeats per minute ≤ 120 and 12 ≤ the number of breaths per minute ≤ 20, then the current physiological state of the human body is a relaxed state.

[0059] The beneficial effects of the present invention:

[0060] First, it uses lidar to detect data of moving objects behind the cyclist (such as cars, bicycles or other pedestrians), and establishes a proximity risk value function through speed, acceleration, relative distance, and volume to quantify the risk of moving objects, enabling the system to accurately perceive the source of danger.

[0061] Second, it introduces traffic complexity (such as traffic congestion level, weather conditions) to further improve the scene adaptability of the danger perception system. Through weighted analysis (combining real-time traffic data and weather data), it improves the scene adaptability of the risk value assessment, that is, it pays more attention to potential dangers in high-complexity scenarios.

[0062] Third, through danger level classification, it designs the first tactile feedback mode to enable users to perceive the threat level of objects behind in real time (such as information on direction, distance, speed, etc.). In cooperation with multi-modal tactile elements (signals such as vibration, piezoelectric, and pyroelectric), it realizes fine-tuning between different types of tactile feedback.

[0063] Fourth, it collects the user's real-time physiological data (heart rate, breathing rate, etc.) through a human physiological sensor to judge the user's current physiological state (nervous, fatigued, relaxed). It adjusts the first tactile feedback mode. In a nervous or fatigued state, it adjusts the tactile feedback mode, simplifies secondary signals or enhances key signals, further improving the user's perception efficiency and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0065] Figure 1 It is the position structure diagram of the tactile feedback element group in the first embodiment.

[0066] Figure 2 It is the method flow chart of the second embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] For the convenience of those skilled in the art to understand, the structure of the present invention will be further described in detail below in combination with the drawings for the embodiments. It should be understood that the steps mentioned in this embodiment, unless specifically stating their order, can be adjusted according to actual needs in their front-back order, and can even be executed simultaneously or partially simultaneously.

[0068] Embodiment 1

[0069] The embodiment of the present invention provides a wearable multi-modal tactile feedback system for cyclists, including:

[0070] A wearable feedback vest, which is sleeved on the upper body of a cyclist and used to generate tactile feedback signals on the upper body of the cyclist;

[0071] A lidar, which is arranged at the rear side of the wearable feedback vest and connected to a controller, and is used to obtain data of moving objects behind the cyclist;

[0072] A human physiological sensor, which is arranged on the wearable feedback vest and connected to a controller, and is used to collect human physiological data;

[0073] Furthermore, the human physiological sensor includes: a heart rate sensor and a respiratory rate sensor;

[0074] The heart rate sensor is arranged at the front chest of the wearable feedback vest and is used to collect the heart rate data of the cyclist;

[0075] The respiratory rate sensor is arranged at the front chest of the wearable feedback vest and is used to collect the respiratory rate data of the cyclist.

[0076] In this embodiment, the heart rate sensor can specifically adopt a photoplethysmogram sensor. An LED light source (generally green light or infrared light) emits light to the skin. Part of the light is absorbed by the skin, and part is reflected or passed through by the blood. A photodetector receives the reflected light and measures its intensity, thereby obtaining a pulse waveform, and the number of heartbeats per minute is obtained through frequency analysis of the pulse signal. The respiratory rate sensor can specifically adopt a piezoelectric sensor. As the chest expands and contracts periodically during inhalation and exhalation, electrical signals of different pressures are generated, and the respiratory cycle and the respiratory rate per minute are obtained through signal analysis.

[0077] Traditional tactile feedback devices often do not consider the physiological states of cyclists or users (such as heart rate changes and respiratory rate changes), and cannot accurately adapt to the needs of users in a dynamic environment. By arranging a heart rate sensor and a respiratory rate sensor at the chest of the vest, using their characteristics of being close to the human core area (the heart and chest cavity), accurate physiological data can be collected in real time to comprehensively monitor the physiological dynamics of users (such as the state of tension, fatigue level, or physiological stress). The chest position fits closely to the heart and chest cavity, with little signal interference and high accuracy for heart rate and respiratory rate. By collecting the heart rate dynamics and respiratory rhythm of users, the physiological state of users (such as energy level, physical stress, health status, etc.) can be comprehensively evaluated for subsequent tactile feedback adjustment.

[0078] A tactile feedback element group, which is distributed at the rear side of the wearable feedback vest and connected to a controller, and is used to perform multi-modal tactile feedback to the cyclist;

[0079] Furthermore, as Figure 1As shown in the figure, the haptic feedback element group includes: a vibration motor group 2, a piezoelectric element group 3, and a thermoelectric element group 4;

[0080] The vibration motor group 2 is respectively arranged at the upper part of the back of the wearable feedback vest 1 and is respectively used for directional haptic feedback of the closest moving object at the rear side;

[0081] The piezoelectric element group 3 is respectively arranged at the middle part of the back of the wearable feedback vest 1 and is respectively used for distance haptic feedback of the closest moving object at the rear side;

[0082] The thermoelectric element group 4 is respectively arranged at the waist of the wearable feedback vest 1 and is respectively used for speed haptic feedback of the closest moving object at the rear side.

[0083] In this embodiment, by reasonably dividing the functional division of piezoelectric + thermoelectric + vibration, it is possible to not only solve the problem of redundant conflict in signal transmission but also reduce the learning cost of haptic feedback. Each feedback method (vibration for direction, thermoelectric for speed, piezoelectric for distance) is always associated with a certain dimension of information, reducing the cognitive burden of the user. Fix the prompt content of each haptic feedback to avoid cognitive errors of the user. The sense of presence is relatively stable and not easily causes obvious interference, and is suitable for transmitting gradually changing relative distances. The physical pressure sense of piezoelectricity is very suitable for simulating the sense of urgency of an object approaching. Speed essentially reflects the movement speed of an object and the sense of threat speed, and the first reaction of users to perceive urgency is often related to the change of temperature (such as cold feeling). The change of cold and warm is easy to trigger strong emotional arousal, which can highlight the speed attribute of an object approaching. The vibration signal has a clear position and is especially suitable for transmitting directional spatial information, and users can immediately locate the source of threat. By binding specific information (distance, speed, direction) to a fixed element combination and maximizing the use of the high sensitivity of different neurons of the user's skin to pressure, temperature, and vibration, an efficient feedback scheme is formed.

[0084] Embodiment Two

[0085] As Figure 1 shown, the embodiment of the present invention provides an interaction method for a wearable multi-modal haptic feedback system for cyclists, including:

[0086] S1 Obtain the data of the moving object at the rear side through the lidar, and construct a proximity danger value function according to the moving object data;

[0087] S101 Obtain the relative distance of the moving object at the rear side through the lidar , speed , acceleration , volume ;

[0088] S102 Through the relative distance of the moving object at the rear side Speed Acceleration Volume Construct a function for approaching the danger value The calculation formula is as follows:

[0089] ,

[0090] Among them, The weight of, The weight of, The weight of, The weight of.

[0091] In this step, speed directly reflects the speed of an object's movement. The greater the speed, the shorter the time for the object to approach the cyclist, and the higher the danger. A threatening object at high speed may pose a greater collision risk to the cyclist. Therefore, it needs to be given key consideration in the calculation of the danger value.

[0092] Acceleration represents the dynamic trend of speed change and is an important physical quantity describing the nature of an object's movement. If the acceleration is positive, it means the object is approaching the cyclist and accelerating, and the danger increases significantly. If the acceleration is negative (deceleration), the threat of the object may gradually decrease. Acceleration enables the system to predict the potential danger of the threatening object in the future for a period of time, rather than just staying at the static analysis of the current speed.

[0093] Distance is the core variable for evaluating approaching danger and has a direct impact on the danger value. The smaller the distance, the greater the danger, because the time required for the object to invade the cyclist's safety space is shorter. Through the distance information, the system can judge whether the threatening object has invaded the "safe distance" range, thus triggering timely warning measures.

[0094] The size of an object's volume determines the possible degree of physical harm. A large volume (such as a car or a truck) poses a significant threat to the cyclist even at a relatively long distance. A small volume (such as a pedestrian or a bicycle) poses a smaller threat to the cyclist, but the danger still needs to be judged in combination with other variables. In a complex traffic environment, introducing the volume factor can distinguish different types of threatening objects and improve the detail and accuracy of the assessment.

[0095] Speed, acceleration, distance, and volume are directly related to the danger, which conforms to the objective physical laws of the potential conflict between moving objects and cyclists. Therefore, these four parameters are selected to construct a function for approaching the danger value 。LiDAR technology obtains the spatial information of surrounding objects by sending pulsed laser beams and receiving reflected signals, and can directly obtain data on multiple moving objects behind the rider through LiDAR; LiDAR emits a laser beam to the target object and receives the laser signal reflected by the target object, calculates the time difference between the laser emission and return (i.e., the propagation time t), and based on the propagation speed of the laser (usually the speed of light c), the relative distance can be calculated; uses multiple frames of LiDAR data to track the position change of the target, and calculates the relative speed in combination with the time interval; acceleration is calculated through the rate of change of the object's speed, so it can be derived based on consecutive multiple frames of LiDAR speed data; LiDAR reconstructs the three-dimensional contour of the target object through multi-point cloud data (Point Cloud) to estimate its volume. LiDAR scanning the environment generates a point cloud model of the object. These point clouds contain the three-dimensional coordinates of different points on the object's surface. According to the point cloud grouping, applying the Bounding Box generation algorithm, the target object is separated and its three-dimensional bounding box (length, width, height) is calculated to estimate the volume. The above calculation methods are all prior arts and will not be elaborated in this invention.

[0096] S2 obtains the traffic complexity value through real-time traffic data and weather data analysis, adjusts the weight of the approaching danger value function through the traffic complexity value, and calculates the approaching danger value;

[0097] S201 obtains vehicle density data, road congestion index data, and average vehicle speed data;

[0098] S202 obtains visibility data, rainfall data, and wind speed data;

[0099] S203 inputs the real-time traffic data and weather data into a pre-trained traffic complexity prediction model to obtain the traffic complexity value C.

[0100] Furthermore, the traffic complexity prediction model is trained through the following steps:

[0101] Collect historical real-time traffic data and weather data, mark the traffic complexity target value according to the historical real-time traffic data and weather data, and construct a training sample;

[0102] Construct a traffic complexity prediction model through the XGBoost model, fit the initial training sample through the decision tree of the XGBoost model, and initialize the traffic complexity prediction value;

[0103] Based on the loss function, calculate the gradient of each sample, and use the gradient and the traffic complexity target value to construct a new decision tree;

[0104] The predicted value of the traffic complexity of the new decision tree and the predicted value of the traffic complexity of the previous round are weighted and summed to iteratively update the traffic complexity prediction model.

[0105] In this step, the traffic complexity dynamically reflects the risk level in different scenarios. By adjusting the weights, the proximity-to-danger value function can be made to more accurately adapt to complex traffic environments. For example, in adverse weather or high congestion situations, more attention is paid to the impact of certain key factors. By constructing a traffic complexity model through multi-dimensional characteristics (such as vehicle density, congestion index, vehicle speed, weather indicators, etc.), the combined impact of traffic and environmental factors can be comprehensively considered.

[0106] Vehicle density data, road congestion index data, average vehicle speed data, etc. can be directly accessed through the map platform, and the existing map platform provides APIs to obtain these real-time traffic data. Visibility data, rainfall data, wind speed data, etc. can be directly accessed through the weather forecast platform to obtain the local weather conditions. These access technologies are existing technologies and will not be elaborated in this invention.

[0107] The XGBoost (Extreme Gradient Boosting) model can handle mixed-type data (numerical and categorical variables) and can construct an accurate prediction model based on the mixed input of real-time traffic data and weather data. Traffic data and weather conditions often have complex non-linear associations (for example, high density + low visibility = high traffic complexity, but this is not a simple linear relationship). Therefore, using the XGBoost model to construct a traffic complexity prediction model can accurately predict the current traffic complexity.

[0108] Updating the traffic complexity in real-time and dynamically can effectively improve the sensitivity and scenario adaptability of the proximity-to-danger value function, especially in a rapidly changing traffic environment. Therefore, by adjusting the weights of the proximity-to-danger value function according to the traffic complexity value and then calculating the proximity-to-danger value, the obtained proximity-to-danger value is the most accurate. Specifically:

[0109] S204 When the traffic complexity value C is greater than 0.5, increase the weight of , the weight of , and the calculation formula is as follows:

[0110] ,

[0111] where is the initial weight of is the initial weight of

[0112] S205 When the traffic complexity value C is less than 0.5, increase The weight of , increase The weight of , and the calculation formula is as follows:

[0113] ,

[0114] Among them, is The initial weight of is The initial weight of

[0115] S206 normalizes and adjusts The weight of , The weight of , The weight of , The weight of to obtain the weights of after the planned adjustment , The weight of , The weight of , The weight of .

[0116] In this step, the higher the traffic complexity value (i.e., C is close to 1), the more attention needs to be paid to the impact of speed and acceleration on danger (i.e., increase and The weights of ). In a high-traffic complexity environment (such as high congestion, low visibility, bad weather, etc.), the traffic flow becomes more unpredictable, and sudden acceleration, hard braking, or gear shifting of vehicles may pose greater safety hazards to cyclists. Speed and acceleration are dynamic variables that directly reflect the drastic changes in the motion state of an object. Therefore, higher attention is required in complex environments. When the traffic complexity is high, for example, on a rainy day or in a situation of high vehicle density, the difficulty of processing visual and auditory information of cyclists increases, and the judgment of distance may not be accurate enough. At this time, more attention needs to be paid to rapidly approaching dynamic threats (such as vehicles approaching at high speed or suddenly accelerating). Speed and acceleration are the indicators that can most quickly reflect these threats. In a high-complexity scenario, focusing only on the relative distance may not be sufficient to quickly recognize threats. For example: A vehicle approaching at high speed from a distance is more dangerous than a vehicle at a close distance but with low speed. Similarly, although large-volume objects are highly dangerous, in a traffic environment dominated by dynamic threats, the threats of speed and acceleration are more direct. In a high-complexity environment, dynamic threats dominate.

[0117] The lower the traffic complexity value (i.e., C is close to 0), the more attention needs to be paid to the impact of relative distance and volume on danger (i.e., increase and In an environment with low traffic complexity (such as low congestion and good weather conditions), the road is more orderly, objects usually move along the expected trajectories, and the fluctuations of dynamic variables (speed and acceleration) are smaller. At this time, relative distance and volume become the main static features of concern. Large-volume objects that are close (such as trucks) can pose a threat to cyclists even if they are stationary at low speeds. In low-complexity scenarios, cyclists have more time to judge distances and paths, so they can pay more attention to static variables (such as which objects are closer and more dangerous). At this time, speed and acceleration are relatively no longer the primary sources of threat. Volume is a direct quantitative indicator of the threat level of an object to the environment. Large-volume objects (such as trucks and buses) are still much more dangerous than small-volume vehicles (such as bicycles) even at low speeds. Therefore, increasing the weight of the volume of approaching objects in low-complexity situations can more reasonably reflect static threats. In a more orderly environment, changes in the speed and acceleration of objects have less impact on the risks faced by cyclists. When dynamic threats decrease, overemphasis on speed and acceleration can easily lead to false alarms or information redundancy. At this time, reducing the weights of speed and acceleration and focusing more on static variables, static threats dominate in low-complexity environments.

[0118] S3 classifies the danger level by approaching the danger value and obtains the first multi-modal tactile feedback pattern according to the danger level;

[0119] S301 classifies those with approaching danger values exceeding the first threshold as high danger levels, provides relative distance feedback of the closest moving object at the rear through a piezoelectric element group, provides speed feedback of the closest moving object at the rear through a thermoelectric element group, and provides direction feedback of the closest moving object at the rear through a vibration motor group;

[0120] S302 classifies those with approaching danger values lower than the first threshold and exceeding the second threshold as medium danger levels, provides relative distance feedback of the closest moving object at the rear through a piezoelectric element group, and provides direction feedback of the closest moving object at the rear through a vibration motor group;

[0121] S303 classifies those with approaching danger values equal to or lower than the second threshold as low danger levels, and provides direction feedback of the closest moving object at the rear through a vibration motor group.

[0122] In this step, the constantly changing distance, speed, and direction of rear moving objects such as cars, bicycles, or pedestrians are the information that cyclists need to pay the most attention to. However, obtaining such data through line of sight or external sensing devices often has delays or blind spots. Distance perception: The piezoelectric element provides real-time feedback on the relative distance, helping the cyclist perceive the proximity of the object. The closer the relative distance between the cyclist and the rear moving object, the stronger the pressure feedback given by the piezoelectric element. Speed perception: The thermoelectric element can indicate the speed of the rear object through temperature changes, especially for dangerous objects approaching at high speed. The faster the speed of the rear moving object, the higher the temperature of the thermoelectric element. Direction perception: The vibration motor provides azimuth feedback, enabling the cyclist to quickly locate the direction of the threat source.

[0123] Compared with providing feedback for multiple approaching objects simultaneously, this solution only targets the "closest" rear object. The dynamic priority adjustment avoids the information chaos and user understanding difficulties caused by multi-signal feedback. In situations with different levels of danger, the information of distance, speed, or direction is mainly used respectively (adjusting the feedback focus according to the danger level), enabling users to focus on the most important information and improving the judgment efficiency.

[0124] Under different danger levels, users' demands for feedback signals are different. For high danger levels, comprehensive and strong feedback (such as distance, speed, and direction) is required to quickly judge the threat and make a response. For medium danger levels: Partial information about the risk (such as distance and direction) is needed to provide sufficient warnings without disturbing operations. For low danger levels: Since the surrounding traffic environment is not complex, only direction prompts are provided to avoid distracting the user's attention from the cycling path. This design solves the problem of how to dynamically adjust the complexity and intensity of feedback signals.

[0125] S4 obtains human physiological data through the human physiological sensor, and adjusts the first multi-modal tactile feedback mode through the human physiological data to obtain the second multi-modal tactile feedback mode.

[0126] S401 analyzes the current physiological state of the human body through heart rate data and respiratory rate data. The current physiological state of the human body includes: tense state, fatigued state, and relaxed state;

[0127] S4011 If the number of heartbeats per minute > 120 and the number of breaths per minute > 20, then the current physiological state of the human body is a tense state;

[0128] S4012 If the number of heartbeats per minute < 60 and the number of breaths per minute < 12, then the current physiological state of the human body is a fatigued state;

[0129] S4013 If 60 ≤ the number of heartbeats per minute ≤ 120 and 12 ≤ the number of breaths per minute ≤ 20, then the current physiological state of the human body is a relaxed state.

[0130] If the current physiological state of the human body is a tense state, then direction feedback of the closest moving object at the rear side is performed through the vibration motor group at all danger levels;

[0131] If the current physiological state of the human body is a fatigued state, then relative distance feedback of the closest moving object at the rear side is performed through the piezoelectric element group at all danger levels, and direction feedback of the closest moving object at the rear side is performed through the vibration motor group;

[0132] If the current physiological state of the human body is a relaxed state, then the first multi-modal tactile feedback mode is retained.

[0133] In this step, in a dangerous environment or an information-intensive scenario, it is easy for the user to be unable to quickly judge the key information due to the excessive complexity or diversity of the tactile signals, resulting in an overly heavy perception burden or reaction delay. When the physiological state is tense (increasing the user's sensitivity), direction feedback is preferentially provided through the vibration motor group, concentrating the tactile information on the most critical hint direction, avoiding interference from additional signals (such as distance or speed) to the user, and only feedbacking the direction of the obstacle in the tense state, so that the user can quickly make a decision and take action. In the fatigued state, the user's sensitivity to tactile signals decreases and it is easy to miss important information. When the physiological state is fatigued, the system will provide two signals: direction feedback (vibration motor group) and distance feedback (piezoelectric element group). By enhancing the perception effect through the dual-modal signals and highlighting the key information (such as the direction and distance of the obstacle) at the same time, the problem of insufficient attention is compensated. By dynamically adapting the feedback mode according to the physiological data, it can help the user quickly process environmental changes.

[0134] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0135] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for realizing the process Figure 1One or more processes and / or blocks Figure 1 Apparatus for the functions specified in one or more blocks

[0136] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction apparatus that implements the functions in the process Figure 1 One or more processes and / or blocks Figure 1 The functions specified in one or more blocks

[0137] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions in the process Figure 1 One or more processes and / or blocks Figure 1 Steps for the functions specified in one or more blocks

[0138] It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim listing several means, several of these means can be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names

[0139] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present invention

[0140] Obviously, those skilled in the art can make various changes and deformations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and deformations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and deformations

[0141] In the present invention, unless otherwise clearly defined or limited, terms such as "installed", "connected", "linked", "fixed", etc. shall be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the internal communication of two components or the interaction relationship between two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0142] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

Claims

1. An interactive method for a wearable multimodal tactile feedback system for a rider, characterized in that: Based on a wearable multimodal tactile feedback system for cyclists, comprising: A wearable feedback vest is mounted on the rider's upper body to generate tactile feedback signals on the rider's upper body; A laser radar is arranged on the rear side of the wearable feedback vest and connected to the controller to obtain data of moving objects behind the rider; A human physiological sensor is disposed on a wearable feedback vest and connected to a controller for collecting human physiological data; The human physiological sensors include: a heart rate sensor and a breathing frequency sensor; The heart rate sensor is arranged at the front chest of the wearable feedback vest to collect the heart rate data of the rider; The breathing rate sensor is arranged at the front chest of the wearable feedback vest to collect the breathing rate data of the rider; A tactile feedback element group is distributed on the rear side of the wearable feedback vest and connected to the controller for providing multi-modal tactile feedback to the rider; The tactile feedback element group includes: a vibration motor group, a piezoelectric element group, and a thermoelectric element group; The vibration motor groups are respectively arranged on the upper back part of the wearable feedback vest, and are respectively used for tactile feedback of the direction of the nearest rear moving object; The piezoelectric element groups are respectively arranged at the middle of the back of the wearable feedback vest, and are respectively used for tactile feedback of the distance of the closest rear moving object; The thermoelectric element groups are respectively arranged at the waist of the wearable feedback vest, and are respectively used for the speed tactile feedback of the closest rear moving object; The method comprises: The laser radar is used to obtain the rear moving object data, and a proximity risk value function is constructed according to the moving object data. Specifically: The relative distance of the moving object behind is obtained by the laser radar ,speed , acceleration ,volume ; Relative distance of the object moving behind ,speed , acceleration ,volume Constructing a function close to the danger value , the calculation formula is as follows: ， in, The weight of The weight of The weight of The weight of The traffic complexity value is obtained by analyzing real-time traffic data and weather data. The weight of the proximity danger value function is adjusted according to the traffic complexity value, and the proximity danger value is calculated. Specifically: When the traffic complexity value C is greater than 0.5, increase Weight , Weight , the calculation formula is as follows: ， in, for The initial weight of for The initial weight of When the traffic complexity value C is less than 0.5, increase Weight ,Increase Weight ; right Weight , Weight , Weight , Weight Perform normalization adjustment to obtain the normalized adjusted Weight , Weight , Weight , Weight ; The danger level is classified by approaching the danger value, and the first multimodal tactile feedback mode is obtained according to the danger level. Specifically: Classify the proximity danger value exceeding the first threshold value as a high danger level, perform relative distance feedback of the closest rear moving object through the piezoelectric element group, perform speed feedback of the closest rear moving object through the thermoelectric element group, and perform direction feedback of the closest rear moving object through the vibration motor group; Classify the proximity danger value lower than the first threshold and higher than the second threshold as a medium danger level, and perform relative distance feedback of the closest rear moving object through the piezoelectric element group, and perform direction feedback of the closest rear moving object through the vibration motor group; Classify the proximity danger value equal to or lower than the second threshold value as a low danger level, and provide feedback on the direction of the nearest rear moving object through the vibration motor group; The human physiological data is obtained through the human physiological sensor, and the first multimodal tactile feedback mode is adjusted according to the human physiological data to obtain the second multimodal tactile feedback mode. Specifically: Analyze the current physiological state of the human body through the heart rate data and the respiratory rate data, wherein the current physiological state of the human body includes: a tense state, a fatigue state, and a relaxed state; If the current physiological state of the human body is a state of tension, the direction feedback of the nearest rear moving object is provided through the vibration motor group at all danger levels; If the current physiological state of the human body is fatigue, the relative distance feedback of the closest rear moving object is performed through the piezoelectric element group at all risk levels, and the direction feedback of the closest rear moving object is performed through the vibration motor group; If the current physiological state of the human body is a relaxed state, the first multimodal tactile feedback mode is retained.

2. The interactive method of a wearable multimodal tactile feedback system for a rider according to claim 1, characterized in that: The traffic complexity value obtained by analyzing real-time traffic data and weather data includes: Obtain vehicle density data, road congestion index data, and average vehicle speed data; Obtain visibility data, rainfall data, and wind speed data; The real-time traffic data and weather data are input into the pre-trained traffic complexity prediction model to obtain the traffic complexity value C.

3. The interactive method of a wearable multimodal tactile feedback system for a rider according to claim 2, characterized in that: The traffic complexity prediction model is trained through the following steps: Collect historical real-time traffic data and weather data, mark traffic complexity target values ​​based on the historical real-time traffic data and weather data, and construct training samples; The traffic complexity prediction model is constructed through the XGBoost model, and the initial training samples are fitted through the XGBoost model decision tree to initialize the traffic complexity prediction value; Based on the loss function, the gradient of each sample is calculated, and a new decision tree is constructed using the gradient and the traffic complexity target value; The traffic complexity prediction value of the new decision tree is weightedly summed with the traffic complexity prediction value of the previous round, and the traffic complexity prediction model is iteratively updated.

4. The interactive method of a wearable multimodal tactile feedback system for a rider according to claim 1, characterized in that: Analyzing the current physiological state of the human body through heart rate data and respiratory rate data includes: If the heart rate is >120 per minute and the breathing rate is >20 per minute, the current physiological state of the human body is tense; If the heart rate is less than 60 per minute and the breathing rate is less than 12 per minute, the current physiological state of the human body is fatigue; If 60≤heartbeats per minute≤120 and 12≤respirations per minute≤20, the current physiological state of the human body is a relaxed state.

Images