Electric wheelchair rocker double-hall collaborative verification position sensing safety processing algorithm and system

Abstract

This invention discloses a safety processing algorithm and system for dual-Hall collaborative verification position sensing of an electric wheelchair joystick, relating to the field of electromechanical control technology. The algorithm includes acquiring a first output signal from a triaxial Hall sensor and a second output signal from a linear Hall sensor, and performing dynamic baseline calibration to obtain two anti-drift position signals. Co-occurrence difference calculation and consistency determination are performed on the two anti-drift position signals to obtain a sensor cluster health status value. Signal fusion and fault reconstruction are performed based on the sensor cluster health status value and the two anti-drift position signals to obtain a fused position command. A confidence-based safety joystick command is generated based on the fused position command and the sensor cluster health status value, and simultaneously sent to the microcontroller module. This provides complete information, including reliability assessment, for subsequent control decisions, thereby constructing a full-link safety protection system from signal perception to control decision-making.

Description

Technical Field

[0001] This invention relates to the field of electromechanical control technology, and in particular to a safety processing algorithm and system for dual Hall effect collaborative verification of position sensing on an electric wheelchair rocker arm. Background Technology

[0002] As an important mobility aid for people with disabilities or limited mobility, the precision and safety of the joystick control in electric wheelchairs directly affect the user's personal safety. Currently, some high-end electric wheelchairs have adopted dual Hall effect sensors (such as a combination of triaxial Hall effect and linear Hall effect) for redundant detection to reduce the risk of single sensor failure and improve the safety level of the device.

[0003] The movement commands of electric wheelchairs rely on the accurate detection of the joystick position sensor. Currently, most products use a single triaxial Hall sensor to obtain the two-dimensional displacement coordinates of the joystick. This single-sensor solution has several hidden failure risks: First, after long-term use, the sensor may experience "center drift" due to component aging, temperature drift, or electromagnetic interference. That is, after the joystick physically returns to center, the zero point of the sensor's output signal shifts and is no longer zero, causing the system to mistakenly believe that there is still an operation command and continue to drive the motor, resulting in a "rollback" phenomenon. Second, the sensor may experience partial functional failure, but the system cannot effectively distinguish it from the normal operation signal, resulting in a serious discrepancy between the user's intention and the wheelchair's actual movement, such as intending to turn left but the wheelchair turning right. Moreover, the structural characteristics of the triaxial Hall sensor also bring about crosstalk problems between the Z-axis and X / Y-axis magnetic field signals. Because the sensing elements inside the sensor cannot be completely superimposed in space, coupled with the influence of the magnetic field at the edge of the magnet, when the joystick is operated at a large angle, the Z-axis signal will be interfered with by the magnetic field components from the X / Y directions, causing the normalization calculation and verification algorithm to fail at certain operating angles. Finally, the zero-position detection accuracy of a single sensor is limited, resulting in a large ambiguity range in joystick return determination, increasing the probability of false alarms. These sensor-level faults are insidious, and the response delays and insufficient safety redundancy caused by joystick sensor failures or main control system malfunctions in electric wheelchairs are difficult to diagnose in real time, constituting a significant potential source of safety accidents. Summary of the Invention

[0004] The technical problem solved by this invention is that electric wheelchairs suffer from center drift, hidden faults, insufficient accuracy and insufficient safety redundancy due to rocker sensor malfunctions, which are difficult to diagnose in real time and constitute an important potential source of safety accidents.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: Firstly, an electric wheelchair rocker arm dual-Hall collaborative verification position sensing safety processing algorithm, comprising:

[0006] Step S1: Acquire the first output signal of the triaxial Hall sensor and the second output signal of the linear Hall sensor, and perform dynamic baseline calibration to obtain two anti-drift position signals;

[0007] Step S2: Calculate the coexistence difference and make a consistency decision on the two anti-drift position signals to obtain the sensor cluster health status value. Based on the sensor cluster health status value and the two anti-drift position signals, perform signal fusion and fault reconstruction to obtain the fused position command.

[0008] Step S3: Obtain the verification threshold according to the fusion position instruction, and perform normalization verification.

[0009] Step S4: Generate a confidence-based safety joystick command based on the fused position command and the sensor cluster health status value, and send it synchronously to the microcontroller module;

[0010] In step S5, after receiving the abnormal signal, the microcontroller module triggers a multi-level alarm mechanism and starts a safety shutdown procedure.

[0011] As a preferred embodiment of the dual Hall effect collaborative verification position sensing safety processing algorithm for the electric wheelchair joystick described in this invention, the microcontroller module is used to receive and respond to the safety joystick command to execute the joystick return-to-center safety monitoring logic. When the confidence level of the safety joystick command is lower than a first preset threshold, an early warning log is triggered.

[0012] As a preferred embodiment of the electric wheelchair rocker dual Hall collaborative verification position sensing safety processing algorithm of the present invention, the microcontroller module is used to receive and respond to the safety rocker command to execute the rocker return-to-center safety monitoring logic, and to use the confidence level of the safety rocker command as an auxiliary safety criterion independent of the microcontroller module's heartbeat mechanism;

[0013] When the confidence level of the safety joystick command is lower than the second preset threshold, the preset monitoring waiting window is skipped, and the safety brake is directly triggered.

[0014] As a preferred embodiment of the dual-Hall collaborative verification position sensing safety processing algorithm for the electric wheelchair rocker arm described in this invention, the algorithm involves sampling the first output signal and the second output signal and performing dynamic baseline calibration to obtain two anti-drift position signals, including:

[0015] When the electric wheelchair system is powered on and the joystick remains within the mechanical zero position range for a preset time, the dynamic baseline calibration process is initiated.

[0016] The first and second output signals are preprocessed using a sliding window filtering algorithm to calculate the updated electrical reference value.

[0017] The updated electrical reference value is compared with the historical reference value stored in non-volatile memory. When the difference between the two exceeds a set percentage, a valid electrical reference value is obtained through multiple consecutive calibrations.

[0018] A piecewise linear interpolation algorithm is used to convert the temperature-compensated Hall voltage characteristic curve into a standardized position signal;

[0019] Calculate the data dispersion of each standardized position signal within the sliding time window, assign calibration confidence weights to each anti-drift position signal based on the comparison result of the data dispersion with the set threshold, and output two anti-drift position signals.

[0020] As a preferred embodiment of the electric wheelchair rocker dual-Hall collaborative verification position sensing safety processing algorithm described in this invention, the following is included: calculating the co-occurrence difference and making a consistency decision on two anti-drift position signals to obtain the sensor cluster health status value, including:

[0021] Calculate the instantaneous difference based on the current sampled values ​​of the two anti-drift position signals;

[0022] Calculate the trend difference based on the changing trends of the two anti-drift position signals within a preset time window;

[0023] Combine instantaneous difference and trend difference into a co-occurrence difference vector;

[0024] If the co-occurrence difference vector satisfies the consistency decision condition, then a weighted decision is made based on the calibration confidence weights of the two anti-drift position signals to obtain the sensor cluster health status value.

[0025] When the health status value of the sensor cluster indicates an abnormal state, the dominant faulty sensor is identified based on the component characteristics of the co-occurrence difference vector, and a corresponding fault type identifier is generated.

[0026] As a preferred embodiment of the electric wheelchair rocker dual Hall collaborative verification position sensing safety processing algorithm of the present invention, the consistency judgment condition is: the instantaneous difference is less than the first set threshold, and the trend difference is less than the second set threshold.

[0027] The first set threshold is positively correlated with the calibration confidence weights of the two anti-drift position signals; the second set threshold is positively correlated with the real-time movement speed of the electric wheelchair.

[0028] As a preferred embodiment of the electric wheelchair rocker dual-Hall collaborative verification position sensing safety processing algorithm of the present invention, wherein: signal fusion and fault reconstruction are performed based on the sensor cluster health status value and two anti-drift position signals to obtain a fused position command, including:

[0029] When the health status value of the sensor cluster is within the preset health status value range, dynamic weighted fusion processing based on calibration confidence weight is performed to obtain a preliminary fused signal;

[0030] When the health status value of the sensor cluster is not within the preset health status value range, the fault sensor isolation and signal reconstruction are performed according to the fault type identifier. The anti-drift position signal that is not identified as a fault sensor is selected as the effective signal source to obtain the reconstructed position signal.

[0031] Gradient amplitude limiting filtering is performed on the preliminary fused signal or reconstructed position signal to limit the instantaneous rate of change of the output signal, thereby obtaining the fused position command.

[0032] As a preferred embodiment of the dual-Hall collaborative verification position sensing safety processing algorithm for the electric wheelchair rocker arm described in this invention, the algorithm includes: performing gradient amplitude limiting filtering on the preliminary fused signal or reconstructed position signal to limit the instantaneous rate of change of the output signal, thereby obtaining the fused position command, including:

[0033] The gradient of the change in the expected command for the current frame is calculated based on the difference between the preliminary fused signal or reconstructed position signal and the fused position command of the previous frame.

[0034] When the change gradient exceeds the maximum permissible change rate of the electric wheelchair system, boundary value limiting processing is performed on the expected command of the current frame based on the maximum permissible change rate.

[0035] The fusion position command for the current frame is calculated based on the expected command for the current frame after the amplitude limiting process and the fusion position command for the previous frame.

[0036] As a preferred embodiment of the dual-Hall collaborative verification position sensing safety processing algorithm for the electric wheelchair joystick described in this invention, the algorithm includes: generating a confidence-based safety joystick command based on the fused position command and the sensor cluster health status value, and synchronously sending it to the microcontroller module and the microcontroller module, including:

[0037] Determine the corresponding confidence value based on the health status value of the sensor cluster, and establish a mapping relationship between the health status value and the confidence value;

[0038] The fusion position command, confidence value, and fault type identifier are combined to generate a safety joystick command data packet;

[0039] The safety joystick command data packets are synchronously sent to the microcontroller module and the microcontroller module through an independent communication channel.

[0040] Secondly, the electric wheelchair rocker arm dual Hall collaborative verification position sensing safety processing system includes a data acquisition module, a compensation module, a judgment module, a response module, and a microcontroller module.

[0041] The acquisition module is used to acquire the first output signal of the triaxial Hall sensor and the second output signal of the linear Hall sensor, and to perform dynamic baseline calibration to obtain two anti-drift position signals;

[0042] The compensation module is used to calculate the coexistence difference and make a consistency decision on the two anti-drift position signals to obtain the sensor cluster health status value, and to perform signal fusion and fault reconstruction based on the sensor cluster health status value and the two anti-drift position signals to obtain the fused position command.

[0043] The determination module is used to obtain the verification threshold according to the fusion position instruction and perform normalization verification;

[0044] The response module is used to generate a confidence-based safety joystick command based on the fused position command and the health status value of the sensor cluster, and send it synchronously to the microcontroller module.

[0045] Upon receiving an abnormal signal, the microcontroller module triggers a multi-level alarm mechanism and initiates a safety shutdown procedure.

[0046] The beneficial effects of this invention are as follows: 1. This invention obtains two anti-drift position signals by sampling the first and second output signals and performing dynamic baseline calibration, effectively suppressing measurement errors caused by temperature drift and component aging; 2. By calculating the coexistence difference and making a consistency decision on the two anti-drift position signals, the health status value of the sensor cluster is obtained, realizing accurate detection and location of early sensor faults; 3. By performing signal fusion and fault reconstruction based on the sensor cluster health status value and the two anti-drift position signals, a fused position command is obtained, ensuring both measurement accuracy under normal conditions and system reliability under fault conditions; 4. By generating a confidence-based safety joystick command based on the fused position command and the sensor cluster health status value, and simultaneously sending it to the microcontroller module, complete information including reliability assessment is provided for subsequent control decisions, thereby constructing a full-link safety protection system from signal perception to control decision-making.

[0047] 2. This invention establishes a dual-redundancy verification system by having the microcontroller module independently execute the joystick return-to-center safety monitoring logic. The microcontroller module triggers an early warning log when the confidence level falls below a first threshold, and directly triggers safety braking when the confidence level falls below a second threshold, achieving hierarchical control of the safety response. This design ensures a low false alarm rate during daily operation and allows for rapid intervention in high-risk scenarios, significantly improving the reliability of safety control for electric wheelchairs.

[0048] 3. This invention generates an anti-drift position signal through dynamic baseline calibration, temperature compensation, piecewise linear interpolation, and sliding window filtering of the dual Hall element voltage signal. By combining the calculation of the co-occurrence difference vector of instantaneous and trend differences, adaptive assessment of the sensor cluster's health status is achieved. This multi-level signal processing mechanism effectively suppresses environmental interference and sensor drift, ensuring continuous accuracy and anti-interference capability of position sensing. This prevents electric wheelchairs from experiencing center drift, latent faults, insufficient accuracy, and insufficient safety redundancy due to rocker sensor malfunctions, thus avoiding safety accidents.

[0049] 4. When the sensor cluster's health status is abnormal, this invention can identify the dominant faulty sensor and generate a fault type identifier. Through dynamic weighted fusion or faulty sensor isolation and reconstruction, combined with gradient amplitude limiting filtering to limit the instantaneous rate of change of the output signal, signal reconstruction and safety control under fault conditions are achieved. This proactive fault management mechanism significantly improves the system's fault tolerance, avoids the risk of loss of control caused by single-point failures, ensures the continuous safe operation of the electric wheelchair, and thus reduces safety accidents. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the basic process of the electric wheelchair rocker dual Hall collaborative verification position sensing safety processing algorithm provided in one embodiment of the present invention.

[0051] Figure 2 This is a flowchart illustrating the normalized verification of a position sensing safety processing algorithm for a dual-Hall collaborative verification of an electric wheelchair rocker arm, as provided in an embodiment of the present invention.

[0052] Figure 3 A schematic diagram of the dynamic S-curve of the electric wheelchair rocker arm dual Hall collaborative verification position sensing safety processing algorithm and system provided in an embodiment of the present invention. Detailed Implementation

[0053] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0054] Example 1, referring to Figures 1-3 This embodiment introduces a dual-Hall collaborative verification position sensing safety processing algorithm for electric wheelchair joysticks, including:

[0055] Step S1: Acquire the first output signal of the triaxial Hall sensor and the second output signal of the linear Hall sensor, and perform dynamic baseline calibration to obtain two anti-drift position signals.

[0056] This invention generates two anti-drift position signals and assigns calibration confidence weights through dynamic baseline calibration, sliding window filtering, temperature compensation, and piecewise linear interpolation. This effectively suppresses environmental interference and sensor drift, ensuring continuous high accuracy and anti-interference capability of the joystick position sensing. The dynamic baseline calibration includes power-on completion, mechanical zero-position timeout, and temperature change trigger conditions.

[0057] Step S2: Calculate the coexistence difference and make a consistency decision on the two anti-drift position signals to obtain the sensor cluster health status value. Based on the sensor cluster health status value and the two anti-drift position signals, perform signal fusion and fault reconstruction to obtain the fused position command.

[0058] This invention constructs a co-occurrence difference vector based on instantaneous difference and trend difference, and combines a first threshold positively correlated with calibration confidence weight and a second threshold positively correlated with real-time motion speed for consistency judgment. This accurately assesses the health status of the sensor cluster and identifies the dominant faulty sensor, improving fault detection sensitivity and positioning accuracy. The sensor cluster health status values ​​are dynamically weighted and fused or fault isolation reconstructed. Gradient amplitude limiting filtering restricts the instantaneous rate of change of the signal, ensuring the stability of the fused signal within the healthy state range. In abnormal states, effective signal sources are reconstructed to ensure control continuity, achieving fault-tolerant operation and safe control of the system.

[0059] Step S3: Obtain the verification threshold according to the fusion position instruction and perform normalization verification.

[0060] Step S4: Generate a confidence-based safety joystick command based on the fused position command and the sensor cluster health status value, and send it synchronously to the microcontroller module;

[0061] In step S5, after receiving an abnormal signal, the microcontroller module triggers a multi-level alarm mechanism and starts a safety shutdown procedure.

[0062] This invention maps health status values ​​to confidence values, generating a safety joystick command data packet containing fused position instructions, confidence levels, and fault indicators. This data packet is then synchronously sent to the main microcontroller module via an independent communication channel, enabling reliable command transmission and collaborative verification with the microcontroller module. This strengthens hierarchical control of safety response and enhances the overall reliability of the system.

[0063] Example 2

[0064] This embodiment describes the implementation steps of the electric wheelchair rocker arm dual Hall collaborative verification position sensing safety processing algorithm, including:

[0065] Step S1: Acquire the first output signal of the triaxial Hall sensor and the second output signal of the linear Hall sensor, and perform dynamic baseline calibration to obtain two anti-drift position signals.

[0066] In this embodiment, preferably, step S1.1: when the electric wheelchair system is powered on and the joystick remains within the mechanical zero position range for a preset time, the dynamic baseline calibration process is initiated.

[0067] This embodiment achieves an intelligent triggering mechanism for the dynamic baseline calibration process through a triple start condition: power-on completion, mechanical zero-position timeout, and temperature change threshold trigger. This ensures that calibration is automatically performed under system initialization, steady-state zero-position, and environmental fluctuation scenarios, effectively adapting to the diverse environmental interference in electric wheelchair usage scenarios and ensuring the real-time accuracy of the position reference.

[0068] Step S1.2: The first and second output signals are preprocessed using a sliding window filtering algorithm to calculate the updated electrical reference value.

[0069] This embodiment uses a sliding window filtering algorithm to preprocess the voltage signal and update the electrical reference value. By smoothing the data within the window, transient noise interference is suppressed, and the stability and reliability of the electrical reference value are improved, laying a high-quality data foundation for the generation of anti-drift position signals.

[0070] Step S1.3: Compare the updated electrical reference value with the historical reference value stored in the non-volatile memory. When the difference between the two exceeds a set percentage, obtain a valid electrical reference value through multiple consecutive calibrations.

[0071] This embodiment ensures the validity verification and dynamic updating of the electrical reference value by comparing the updated electrical reference value with the historical reference value and by performing multiple consecutive calibrations when the difference exceeds the threshold. This avoids reference drift caused by a single calibration error and enhances the accuracy guarantee of position signal generation.

[0072] Step S2: Calculate the coexistence difference and make a consistency decision on the two anti-drift position signals to obtain the sensor cluster health status value. Based on the sensor cluster health status value and the two anti-drift position signals, perform signal fusion and fault reconstruction to obtain the fused position command.

[0073] In this preferred embodiment, step S2.1: calculate the instantaneous difference based on the current sampled values ​​of the two anti-drift position signals.

[0074] This embodiment calculates the instantaneous difference between the current sampled values ​​of two anti-drift position signals to achieve real-time synchronous difference monitoring of the joystick position signal, providing an instantaneous state quantification index for the coexistence difference vector and supporting a fast-response anomaly detection mechanism.

[0075] Step S2.2: Calculate the trend difference based on the changing trends of the two anti-drift position signals within a preset time window.

[0076] This embodiment calculates the trend difference of signal change within a preset time window, captures the dynamic evolution characteristics of sensor signals, identifies potential trend-based fault modes, and improves the lead time and accuracy of fault prediction.

[0077] Step S2.3: Combine the instantaneous difference and trend difference into a co-occurrence difference vector.

[0078] This embodiment combines instantaneous difference degree and trend difference degree into a co-existing difference degree vector, constructs a unified representation form for the fusion of multi-dimensional difference features, realizes the comprehensive quantification of instantaneous state and trend evolution, and provides a structured data foundation for consistency judgment.

[0079] Step S2.4: If the co-occurrence difference vector satisfies the consistency decision condition, then a weighted decision is made based on the calibration confidence weights of the two anti-drift position signals to obtain the sensor cluster health status value.

[0080] This embodiment uses a first set threshold positively correlated with the calibration confidence weight and a second set threshold positively correlated with the real-time motion speed as consistency decision conditions to realize a dynamic adaptive threshold adjustment mechanism, ensuring that the health status of the sensor cluster can be accurately determined under different confidence levels and motion speeds, and enhancing the working condition adaptability of the decision conditions.

[0081] In this embodiment, the consistency judgment condition is: the instantaneous difference is less than a first set threshold, and the trend difference is less than a second set threshold; wherein, the first set threshold is positively correlated with the calibration confidence weight of the two anti-drift position signals; and the second set threshold is positively correlated with the real-time movement speed of the electric wheelchair.

[0082] Step S2.5: When the health status value of the sensor cluster indicates an abnormal state, identify the dominant faulty sensor based on the component characteristics of the co-occurrence difference vector and generate the corresponding fault type identifier.

[0083] When the sensor cluster is in an abnormal health state, this embodiment identifies the dominant faulty sensor based on the component features of the co-occurrence difference vector and generates a fault type identifier, thereby achieving accurate location of the fault source and clear identification of the fault type. This provides a direct basis for subsequent fault isolation, signal reconstruction and maintenance decisions, and improves the efficiency and pertinence of system fault handling.

[0084] Based on the sensor cluster health status values ​​and two anti-drift position signals, signal fusion and fault reconstruction are performed to obtain the fused position commands, including:

[0085] When the health status value of the sensor cluster is within the preset health status value range, dynamic weighted fusion processing based on calibration confidence weights is performed to obtain a preliminary fused signal.

[0086] When the health status value of the sensor cluster is within a preset health range, dynamic weighted fusion processing based on calibration confidence weights is performed to obtain a preliminary fused signal. This embodiment combines the calibration confidence weights of each anti-drift position signal for weighted decision-making, making full use of the effective information from high-confidence sensors in a healthy state, improving the accuracy and stability of the fused signal, ensuring the accuracy and consistency of joystick position sensing, and providing highly reliable position input for safety control.

[0087] When the health status value of the sensor cluster is not within the preset health status value range, fault sensor isolation and signal reconstruction are performed according to the fault type identifier. The anti-drift position signal that is not identified as a fault sensor is selected as the effective signal source to obtain the reconstructed position signal.

[0088] When the health status value of the sensor cluster exceeds the preset health range, faulty sensor isolation and signal reconstruction are performed according to the fault type identifier. An anti-drift position signal not identified as faulty is selected as the valid signal source to obtain the reconstructed position signal. This embodiment accurately identifies and isolates faulty sensors and selects valid signal sources for reconstruction, maintaining basic system control functions in fault scenarios, avoiding system failure due to single-point failures, and significantly improving the fault-tolerant operation capability and safety redundancy of the electric wheelchair.

[0089] Gradient amplitude limiting filtering is performed on the preliminary fused signal or reconstructed position signal to limit the instantaneous rate of change of the output signal, thereby obtaining the fused position command.

[0090] The gradient of the change in the expected command for the current frame is calculated based on the difference between the preliminary fused signal or reconstructed position signal and the fused position command of the previous frame.

[0091] This embodiment achieves quantitative monitoring of signal changes by calculating the gradient of the difference between the expected command in the current frame and the fused position command in the previous frame, providing precise dynamic parameter support for instantaneous rate of change control.

[0092] When the change gradient exceeds the maximum allowable change rate of the electric wheelchair system, the expected command of the current frame is subjected to boundary value limiting processing based on the maximum allowable change rate.

[0093] When the gradient change exceeds the system's maximum allowable rate of change, boundary value limiting processing is performed on the expected command of the current frame based on the maximum allowable rate of change. This effectively suppresses signal abrupt changes, prevents dangerous operations caused by sudden joystick commands, and improves the stability and safety of system operation.

[0094] The fusion position command for the current frame is calculated based on the expected command for the current frame after the amplitude limiting process and the fusion position command for the previous frame.

[0095] This embodiment calculates the fused position command of the current frame based on the expected command of the current frame after amplitude limiting and the fused position command of the previous frame, ensuring smooth signal transition and continuity, avoiding command jumps caused by amplitude limiting, ensuring the continuity and trackability of control commands, and further improving the operational safety and riding comfort of electric wheelchairs.

[0096] Step S3: Obtain the verification threshold according to the fusion position instruction and perform normalization verification.

[0097] In this preferred embodiment, step S31: a correspondence between the fusion position command and the dynamic threshold is established in advance to form a dataset, and the dynamic threshold in the dataset corresponds one-to-one with the range of control parameters in the fusion position command. The verification threshold is matched in the dataset using a lookup table method.

[0098] Step S32: Calculate the offset of the triaxial Hall sensor and the linear Hall sensor, specifically including:

[0099] Triaxial Hall offset O_triaxial = K1 × (Bz - Bz0);

[0100] Where K1 represents the calibration coefficient; Bz0 represents the Z-axis magnetic field strength at the center, Bz represents the Z-axis magnetic field strength, "triaxial" is a compound word composed of the prefixes "tri-" (three) and "axial" (axis), literally translated as "three-axis", and O is an abbreviation for Offset. Therefore, O_triaxial represents the three-axis Hall offset;

[0101] Linear Hall offset O_linear = K2 × (Vout - V0);

[0102] Where K2 represents the sensitivity coefficient, V0 represents the center point voltage, and Vout is an analog voltage signal that is proportional to the axial displacement.

[0103] Step S33: Normalization check includes converting the offset to a percentage of full scale:

[0104] O_tri_norm = (O_triaxial / FullScale_Tri) × 100%;

[0105] O_lin_norm = (O_linear / FullScale_Linear) × 100%;

[0106] Delta = |O_tri_norm - O_lin_norm|;

[0107] Wherein, O_tri_norm represents the normalized three-axis Hall offset; O_lin_norm represents the normalized linear Hall offset; FullScale_Tri represents the offset of the three-axis Hall full scale; and FullScale_Linear represents the offset of the linear Hall full scale.

[0108] Anomaly detection: If Delta > verification threshold ε (ε=2%), a location sensing anomaly is triggered.

[0109] Furthermore, triaxial Hall center drift diagnosis is performed, which includes:

[0110] The trigger condition is set to the joystick returning to center (O_linear ≈ 0).

[0111] The fault determination is set to be based on the following conditions:

[0112] |O_triaxial| > 1.5% × FullScale_Tri;

[0113] If the abnormality persists for more than 50 milliseconds, it is determined to be a three-axis Hall zero-point drift fault.

[0114] By normalizing the verification, we can achieve sensitive detection of sensor faults and signal anomalies, avoid the failure to detect faults in time due to the hidden nature of sensor-level faults, and solve the biggest safety hazard of single point of failure.

[0115] Step S4: Generate a confidence-based safety joystick command based on the fused position command and the sensor cluster health status value, and send it synchronously to the microcontroller module and the microcontroller module.

[0116] In this preferred embodiment, step S4.1: determine the corresponding confidence value based on the health status value of the sensor cluster, and establish a mapping relationship between the health status value and the confidence value.

[0117] This embodiment establishes a direct mapping between the health status values ​​and confidence values ​​of the sensor cluster, enabling a quantitative characterization of the reliability of safety joystick commands. This mechanism allows the main microcontroller module to evaluate the effectiveness of commands based on explicit confidence indicators, dynamically adjust safety response strategies when health status fluctuates, and enhance the transparency and traceability of system safety decisions.

[0118] Step S4.2: Combine the fused position command, confidence value, and fault type identifier to generate a safety joystick command data packet.

[0119] This embodiment integrates position commands, confidence values, and fault type identifiers into a safety joystick command data packet, forming a complete command set containing position information, reliability indicators, and fault diagnosis information. This structured data packet design ensures that the main microcontroller module synchronously receives multi-dimensional decision-making data, supporting closed-loop control throughout the entire process of command validity verification, fault isolation decision-making, and safety braking triggering during collaborative verification.

[0120] Step S4.3: Synchronously send the safety joystick command data packet to the microcontroller module and the microcontroller module through an independent communication channel.

[0121] This embodiment employs an independent communication channel to synchronously send safety joystick command data packets to the main microcontroller module, constructing a physically isolated redundant transmission path. This design avoids command loss or delay caused by single-channel failure, ensuring the real-time performance and integrity of command transmission even under abnormal scenarios such as electromagnetic interference and channel congestion. It supports the efficient execution of the hierarchical safety response mechanism, ultimately improving the overall safety redundancy and operational reliability of the electric wheelchair system.

[0122] In this embodiment, the microcontroller module is used to receive and respond to the safety joystick command to execute the joystick return-to-center safety monitoring logic. When the confidence level of the safety joystick command is lower than a first preset threshold, an early warning log is triggered.

[0123] In step S5, after receiving the abnormal signal, the microcontroller module triggers a multi-level alarm mechanism and starts a safety shutdown procedure.

[0124] In this preferred embodiment, the microcontroller module is used to receive and respond to the safety joystick command to execute the joystick return-to-center safety monitoring logic, and uses the confidence level of the safety joystick command as an auxiliary safety criterion independent of the microcontroller module's heartbeat mechanism; when the confidence level of the safety joystick command is lower than a second preset threshold, a preset monitoring waiting window is crossed, and safety braking is directly triggered. The microcontroller module adopts a dynamic S-curve safety stop algorithm to control the electric wheelchair to stop smoothly, transforming the fault detection result into maximum protection for the user and minimizing the risk of secondary injury.

[0125] Compared to related technologies that only focus on detection and ignore downtime safety, or that only use a single / homogeneous sensor, this embodiment achieves a qualitative leap in active safety, fault diagnosis depth, and humanized failure protection, providing electric wheelchair users with a higher level of safety assurance and demonstrating significant technological advancement and market differentiation competitive advantages.

[0126] Example 3, as Figure 2 As shown in the figure, this embodiment introduces an electric wheelchair rocker arm dual Hall collaborative verification position sensing safety processing system, including:

[0127] The system includes a data acquisition module, a compensation module, a decision-making module, a response module, and a microcontroller module.

[0128] The acquisition module is used to acquire the first output signal of the triaxial Hall sensor and the second output signal of the linear Hall sensor, and to perform dynamic baseline calibration to obtain two anti-drift position signals;

[0129] The compensation module is used to calculate the coexistence difference and make a consistency decision on the two anti-drift position signals to obtain the sensor cluster health status value, and to perform signal fusion and fault reconstruction based on the sensor cluster health status value and the two anti-drift position signals to obtain the fused position command.

[0130] The determination module is used to obtain the verification threshold according to the fusion position instruction and perform normalization verification;

[0131] The response module is used to generate a confidence-based safety joystick command based on the fused position command and the health status value of the sensor cluster, and send it synchronously to the microcontroller module.

[0132] Upon receiving an abnormal signal, the microcontroller module triggers a multi-level alarm mechanism and initiates a safety shutdown procedure.

[0133] The implementation steps of the above modules are the same as those in Embodiment 1 or 2, and will not be repeated here.

[0134] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. 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. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0135] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A dual-Hall collaborative verification position sensing safety processing algorithm for electric wheelchair rocker arm, characterized in that, include: Step S1: Acquire the first output signal of the triaxial Hall sensor and the second output signal of the linear Hall sensor, and perform dynamic baseline calibration to obtain two anti-drift position signals; Step S2: Calculate the coexistence difference and make a consistency decision on the two anti-drift position signals to obtain the sensor cluster health status value. Based on the sensor cluster health status value and the two anti-drift position signals, perform signal fusion and fault reconstruction to obtain the fused position command. Step S3: Obtain the verification threshold according to the fusion position instruction, and perform normalization verification; Step S4: Generate a confidence-based safety joystick command based on the fused position command and the sensor cluster health status value, and send it synchronously to the microcontroller module; In step S5, after receiving the abnormal signal, the microcontroller module triggers a multi-level alarm mechanism and starts a safety shutdown procedure.

2. The electric wheelchair rocker arm dual Hall effect collaborative verification position sensing safety processing algorithm as described in claim 1, characterized in that, The microcontroller module is used to receive and respond to the safety joystick command to execute the joystick return-to-center safety monitoring logic. When the confidence level of the safety joystick command is lower than a first preset threshold, an early warning log is triggered.

3. The electric wheelchair rocker arm dual Hall effect collaborative verification position sensing safety processing algorithm as described in claim 1, characterized in that, The microcontroller module is used to receive and respond to the safety joystick command to execute the joystick return-to-center safety monitoring logic, and to use the confidence level of the safety joystick command as an auxiliary safety criterion independent of the microcontroller module's heartbeat mechanism; When the confidence level of the safety joystick command is lower than the second preset threshold, the preset monitoring waiting window is skipped, and the safety brake is directly triggered.

4. The electric wheelchair rocker arm dual Hall collaborative verification position sensing safety processing algorithm as described in claim 1, characterized in that, Two anti-drift position signals are obtained by sampling the first and second output signals and performing dynamic baseline calibration, including: When the electric wheelchair system is powered on and the joystick remains within the mechanical zero position range for a preset time, the dynamic baseline calibration process is initiated. The first and second output signals are preprocessed using a sliding window filtering algorithm to calculate the updated electrical reference value. The updated electrical reference value is compared with the historical reference value stored in non-volatile memory. When the difference between the two exceeds a set percentage, a valid electrical reference value is obtained through multiple consecutive calibrations.

5. The electric wheelchair rocker arm dual Hall effect collaborative verification position sensing safety processing algorithm as described in claim 4, characterized in that, The health status value of the sensor cluster is obtained by calculating the co-occurrence difference and determining the consistency between two anti-drift position signals, including: Calculate the instantaneous difference based on the current sampled values ​​of the two anti-drift position signals; Calculate the trend difference based on the changing trends of the two anti-drift position signals within a preset time window; Combine instantaneous difference and trend difference into a co-occurrence difference vector; If the co-occurrence difference vector satisfies the consistency decision condition, then a weighted decision is made based on the calibration confidence weights of the two anti-drift position signals to obtain the sensor cluster health status value. When the health status value of the sensor cluster indicates an abnormal state, the dominant faulty sensor is identified based on the component characteristics of the co-occurrence difference vector, and a corresponding fault type identifier is generated.

6. The electric wheelchair rocker arm dual Hall effect collaborative verification position sensing safety processing algorithm as described in claim 5, characterized in that, The consistency judgment condition is: the instantaneous difference is less than a first set threshold, and the trend difference is less than a second set threshold; The first set threshold is positively correlated with the calibration confidence weights of the two anti-drift position signals; the second set threshold is positively correlated with the real-time movement speed of the electric wheelchair.

7. The electric wheelchair rocker arm dual Hall effect collaborative verification position sensing safety processing algorithm as described in claim 6, characterized in that, Based on the sensor cluster health status values ​​and two anti-drift position signals, signal fusion and fault reconstruction are performed to obtain fused position commands, including: When the health status value of the sensor cluster is within the preset health status value range, dynamic weighted fusion processing based on calibration confidence weight is performed to obtain a preliminary fused signal; When the health status value of the sensor cluster is not within the preset health status value range, the fault sensor isolation and signal reconstruction are performed according to the fault type identifier. The anti-drift position signal that is not identified as a fault sensor is selected as the effective signal source to obtain the reconstructed position signal. Gradient amplitude limiting filtering is performed on the preliminary fused signal or reconstructed position signal to limit the instantaneous rate of change of the output signal, thereby obtaining the fused position command.

8. The electric wheelchair rocker arm dual Hall effect collaborative verification position sensing safety processing algorithm as described in claim 7, characterized in that, Gradient amplitude limiting filtering is performed on the preliminary fused signal or reconstructed position signal to limit the instantaneous rate of change of the output signal, resulting in a fused position command, including: The gradient of the change in the expected command for the current frame is calculated based on the difference between the preliminary fused signal or reconstructed position signal and the fused position command of the previous frame. When the change gradient exceeds the maximum permissible change rate of the electric wheelchair system, boundary value limiting processing is performed on the expected command of the current frame based on the maximum permissible change rate. The fusion position command for the current frame is calculated based on the expected command for the current frame after the amplitude limiting process and the fusion position command for the previous frame.

9. The electric wheelchair rocker arm dual Hall effect collaborative verification position sensing safety processing algorithm as described in claim 8, characterized in that, Based on the fused location command and the sensor cluster health status value, a confidence-based safety joystick command is generated and synchronously sent to the microcontroller module and the microcontroller module, including: Determine the corresponding confidence value based on the health status value of the sensor cluster, and establish a mapping relationship between the health status value and the confidence value; The fusion position command, confidence value, and fault type identifier are combined to generate a safety joystick command data packet; The safety joystick command data packets are synchronously sent to the microcontroller module and the microcontroller module through an independent communication channel.

10. A dual-Hall collaborative verification position sensing safety processing system for an electric wheelchair joystick, the system being used to execute the dual-Hall collaborative verification position sensing safety processing algorithm for an electric wheelchair joystick as described in any one of claims 1-9, characterized in that, It includes a data acquisition module, a compensation module, a decision-making module, a response module, and a microcontroller module; The acquisition module is used to acquire the first output signal of the triaxial Hall sensor and the second output signal of the linear Hall sensor, and to perform dynamic baseline calibration to obtain two anti-drift position signals; The compensation module is used to calculate the coexistence difference and make a consistency decision on the two anti-drift position signals to obtain the sensor cluster health status value, and to perform signal fusion and fault reconstruction based on the sensor cluster health status value and the two anti-drift position signals to obtain the fused position command. The determination module is used to obtain the verification threshold according to the fusion position instruction and perform normalization verification; The response module is used to generate a confidence-based safety joystick command based on the fused position command and the health status value of the sensor cluster, and send it synchronously to the microcontroller module. Upon receiving an abnormal signal, the microcontroller module triggers a multi-level alarm mechanism and initiates a safety shutdown procedure.

Images