A double-car type adaptive intelligent riding light control system with six-axis IMU, no GPS slope solution, dual-mode power calculation and star flash networking

Abstract

The application discloses a double-bicycle type intelligent riding light control system with six-axis IMU, no-GPS slope calculation, dual-mode power calculation and star flash networking, and belongs to the field of non-motor vehicle riding accessories. The system comprises a handlebar sensor, a millimeter wave radar helmet, a central control, a six-axis IMU, star flash communication, light voice and a power module. The central control realizes no-GPS slope calculation by means of the IMU, and the GPS can be used for correcting temperature drift; formula and torque sensors realize dual-mode power calculation; double-parallel deceleration determination avoids slope false triggering; star flash realizes 500-meter multi-device networking, and synchronizes light, data and rescue relay. The system is suitable for multiple riding scenes, integrates radar early warning, dumping rescue and anti-theft fault diagnosis, supports parameter customization, works offline, is waterproof and compatible with human-powered and electric bicycles. The application solves the defects of traditional riding lights, such as dependence on GPS, short communication distance, fixed parameters and single vehicle type adaptation, and meets the safety requirements of commuting and training.

Description

Technical Field

[0001] This invention belongs to the field of intelligent cycling safety accessories technology for non-motorized vehicles. Specifically, it involves an intelligent cycling lighting control system that integrates multi-node six-axis IMU attitude sensing, independent slope calculation without GPS, dual-mode cycling power calculation, multi-device networking collaboration within 500 meters, adaptive lighting linkage for five cycling scenarios, unified management of batteries for multiple accessories, and multi-dimensional fault and mechanical diagnostic warnings. It is also compatible with both human-powered bicycles and electric bicycles, and comes with Android, iOS, and HarmonyOS mobile apps to enable full parameter customization and motion data visualization. Background Technology

[0002] Current non-motorized vehicle riding warning lights on the market have nine major defects, failing to cover the needs of all riding scenarios, including commuting, road training, mountain biking, long-distance riding, and multi-person team riding: 1. The deceleration recognition is limited to mechanical braking and lacks recognition for 20-second rolling speed coasting deceleration. Gravity on the slope causes a large number of false triggers. 2. The gradient calculation relies entirely on GPS; there is no signal failure in tunnels or indoor cycling platforms, and there is no independent IMU gradient calculation capability. 3. Lacks a complete cycling power system, lacks both standard software estimation and high-end torque measurement solutions, and lacks FTP and TSS training load calculations; 4. It lacks multi-scenario adaptive strategies, cannot recognize road conditions, fatigue, or poor riding posture, and lacks rear radar graded warning, automatic emergency call upon tipping over, vehicle vibration anti-theft, and chain / wheelset / bearing mechanical fault diagnosis. 5. Communication is limited to Bluetooth point-to-point, with no satellite flash for long-distance networking. Teams cannot share power or gradient, and the leader's braking and steering cannot synchronize with the team's lights. There is no in-team intercom or SOS repeater. 6. Battery level is only displayed as a percentage, and the actual battery life cannot be calculated based on historical power consumption over multiple days. The batteries of various peripherals are scattered and there is no unified summary page. 7. All threshold, duration, and resistance parameters are fixed and cannot be customized, resulting in narrow patent protection. 8. Hardware disconnected from the phone results in numerous functional failures, lack of local algorithm processing and offline timing caching, and the device lacks encrypted hierarchical account binding, making it susceptible to code theft. 9. The charging method is limited, there is no whole-device waterproofing or multiple power hardware protections, the battery life is severely reduced in low temperatures, and there is no real-time dynamic power consumption scheduling.

[0003] The industry urgently needs an integrated cycling lighting system that combines IMU-based slope tracking without GPS, dual-mode power, multi-device networking with star-flash functionality, dual deceleration detection, five-scenario adaptive capabilities, full parameter customization, unified management of multiple battery components, mechanical fault diagnosis, and dynamic power consumption scheduling.

[0004] The industry urgently needs an integrated cycling lighting system that combines IMU-based slope tracking without GPS, dual-mode power, multi-device networking with star-flash functionality, dual deceleration detection, five-scenario adaptive capabilities, full parameter customization, unified management of multiple battery components, mechanical fault diagnosis, and dynamic power consumption scheduling. Summary of the Invention

[0005] 1. Purpose of the invention To address the shortcomings of existing technologies, such as easy accidental activation of turn signals, lack of deceleration warning during coasting, reliance on GPS without independent slope measurement, lack of quantification of riding power, lack of multi-device networking, poor scene adaptation, rudimentary power management, fixed parameters, reliance on mobile phones, insufficient protective power supply, and lack of vehicle mechanical diagnostics, this invention provides a dual-model adaptive intelligent riding lighting control system with a six-axis IMU, GPS-free slope calculation, dual-mode power measurement, and star-flash networking.

[0006] This invention achieves 10 core innovative goals: 1) The six-axis IMU calculates road slope purely in hardware without GPS. It continuously outputs slope in tunnels and on the cycling trainer. GPS corrects temperature drift online. The slope is used as the core input for power, lighting, and elevation statistics. 2) Dual-mode cycling power: The standard version software uses a three-stage resistance formula for estimation, and a high-end torque sensor for high-precision measurement. It automatically calculates FTP, five power ranges, and TSS training load. 3) Two parallel deceleration determination methods: instantaneous deceleration and 20-second rolling average vehicle speed, covering both braking and coasting deceleration. All parameters are either default or customizable. 4) StarFlash SLE defaults to 500-meter multi-device networking, team power / gradient / vehicle speed sharing, leader lighting synchronization, intra-team bone conduction intercom, and distress message relay forwarding; 5) Complete adaptive logic for five major cycling scenarios: commuting radar warning and anti-theft, road riding posture correction and fatigue reminder, off-road resistance automatic correction, long-distance off-road route book and elevation statistics, and team linkage; 6) IMU vibration detection identifies wear on chains, wheelsets, and bearings, and outputs vehicle maintenance diagnostic reminders; 7) Abandoning battery percentage, the system defaults to calculating battery life based on 5 days of historical power consumption, and supports 3 / 7 day quick and customizable durations. A dedicated page in the central control unit / app summarizes the total usage hours of all accessories. 8) All numerical parameters are unified with "default preferred value + user-defined value", which expands the scope of patent protection and avoids competitors from simply changing parameters to bypass protection; 9) BLE5.3 + StarFlash dual-mode AES encryption three-level account binding, offline time-series caching, hardware can run completely independently without the APP; 10) The entire unit is IPX6 waterproof, features dual charging, multiple power protections, dynamic peripheral refresh rate scheduling based on real-time riding power, low-temperature power consumption compensation, and compatibility with both human-powered and electric bicycles. 2. Overall System Hardware Architecture

[0007] Hardware list: Handlebar integrated sensor module, smart cycling helmet module, STM32 embedded central control host, distributed six-axis IMU unit, wheel speed Hall effect sensor, cadence Hall effect sensor, photosensitive sensor, rear millimeter-wave radar, multiple lights, bone conduction voice module, BLE5.3 + StarSpark SLE dual-mode communication, three-terminal APP, rechargeable power supply, waterproof charging interface; the central control unit has a reserved ADC expansion interface for the bottom bracket / crank torque sensor. The torque sensor is a professional high-end optional accessory and does not need to be added to the standard version.

[0008] Hardware division of labor: 1) Handlebar integrated sensor module: Built-in six-axis IMU, wheel speed Hall effect sensor, steering data acquisition, and star flash unit, which uploads steering, speed, and raw attitude data; 2) Smart helmet module: millimeter-wave radar, helmet six-axis IMU, bone conduction voice, impact acquisition, tail warning light, to realize rear detection, posture acquisition, voice alarm and collision recognition; 3) STM32 central control: The only core computing unit, fully carrying all the underlying algorithms for IMU slope calculation, dual-mode power formulas, road condition recognition, scene linkage, power consumption scheduling, and fault diagnosis; 4) Dual-mode communication: BLE5.3 is responsible for sending mobile phone pairing parameters; StarShine SLE enables low-latency networking within 500 meters and high-speed data synchronization among multiple devices; 5) Three-terminal APP: parameter customization, power gradient report, offline line manual, alarm push, account permissions, cloud log export; 6) Power module: Overall power supply SOC detection, power consumption statistics of multiple components, historical battery life calculation, dynamic power consumption scheduling, and low temperature compensation; 7) Charging interface: magnetic / contact dual mode, IPX6 waterproof rating, multiple hardware protections. 3. Six-axis IMU without GPS slope calculation module

[0009] A six-axis IMU consists of a three-axis accelerometer and a three-axis gyroscope. 1) The static gravity component is extracted from the accelerometer to coarsely solve the pitch angle, and the slope of the basic road surface is obtained; 2) Gyroscope angular velocity integral compensation for bumps and vibrations, smoothing out slope values ​​and eliminating jumps; 3) The entire route does not rely on GPS / altitude; it provides a stable gradient output even in tunnels, dense forests, underground parking garages, and indoor cycling platforms. 4) When the GPS signal is normal, the long-term temperature drift cumulative error of the IMU with altitude correction is integrated to balance instantaneous accuracy and long-term stability; the slope output is used for: power calculation, road condition recognition, headlight brightness adjustment, long-distance cumulative ascent / descent mileage, and steep slope prediction off-route. 4. Dual-mode riding power calculation unit

[0010] Mode 1: Standard software estimation (no additional hardware, standard configuration) Data collection includes vehicle speed v, IMU slope θ, and APP-defined total weight of people and vehicle m, tire rolling resistance coefficient Crr, frontal area CdA, and transmission efficiency η; Three-stage resistance power formula: Rolling resistance power: P_roll = m・g・Crr・v Power generated by gravity when climbing a slope: Pslope = m・g・v・sinθ Air resistance power: P_wind = 0.5・ρ・CdA・v³ Total resistance power: Ptotal = Proll + Pslope + Pwind Rider's estimated output power: P_estimated = P_total / η The system automatically identifies and corrects large Crr errors when switching between off-road gravel conditions; it outputs instantaneous / average / peak power, total work done, and calories in real time; it automatically calculates the FTP threshold at a constant speed for 20 minutes, divides the power range into five levels, and calculates the TSS load for a single training session. Mode 2: High-power measured output (with optional torque sensor)

[0011] The system collects pedaling torque T, converts pedal frequency to crank angular velocity ω, and measures actual power P: Actual Power = T・ω, with a measurement accuracy far exceeding software estimation. A dedicated ADC acquisition channel is reserved in the central control unit for plug-and-play mode switching. Power data is synchronized with the central control unit via a flash camera and cached, then used by the app to generate segmented motion curve reports. 5. Smooth deceleration dual-parallel decision logic

[0012] The two sets of logic operate synchronously. If either condition is met to trigger the deceleration lights, the mechanical brakes, with the highest priority, will directly disable the deceleration lights. 1) Instantaneous deceleration determination: Default 0.3~1.2m / s² gentle, 1.2~2.5m / s² moderate, >2.5m / s² emergency flash, all threshold values ​​can be customized; IMU pitch angle dynamically corrects the slope threshold, automatic offset on uphill and downhill slopes eliminates gravity misjudgment; emergency braking simultaneously verifies the dual conditions of sudden speed drop + high deceleration; 2) Rolling average vehicle speed determination: The default is a 20-second statistics window and a default 5% vehicle speed reduction. Both parameters can be customized. Based on the cruise speed, if the 20-second average vehicle speed reduction exceeds the threshold, the accelerator is released to coast and decelerate. The low-frequency warning is automatically lit without needing to squeeze the brake. 6. Three-level lighting priority and linkage mode

[0013] Priority from highest to lowest: emergency flashing light for fall > brake lights > turn signals; three working modes: brake priority (default), turn signals and brake lights coexisting, and emergency flashing light for fall with the highest priority; the fall detection angle and duration are default parameters, but can be customized; for multi-person teams, the leader's turn signal / brake signal is synchronized, and the lights of the entire team are brightened synchronously; the system automatically and dynamically adjusts the light brightness and flashing frequency according to real-time power, slope, and road conditions. 7. Five-scenario adaptive full suite of functions Scenario 1: Urban commuting by bicycle

[0014] IMU vibration and slope recognition for paved roads, speed bumps, and potholes; automatic adjustment of headlight brightness and flashing; full-lighting tunnels; millimeter-wave radar fusion with vehicle speed and slope classification, helmet vibration + bone conduction + rear headlight warning; collision and tipping local cache timing, automatic SOS SMS with GPS, message retransmission without GPS cache; low-power vibration monitoring of stationary vehicle IMU, audible and visual anti-theft alarm for abnormal shaking; segmented statistics of physical exertion on flat roads / uphill slopes, generating weekly commuting reports; IMU identification of abnormal vibration in chains, wheelsets, and bearings, pushing maintenance reminders. Scene 2: Highway Professional Training

[0015] It combines real-time power and gradient bone conduction to announce the optimal cadence; low cadence with high load prompts to adjust gear ratio; FTP and TSS training load are automatically calculated after 20 minutes of constant speed; the helmet and handlebar dual IMU fusion recognition identifies poor riding posture such as forward leaning, center of gravity shift, and shoulder shrugging, and provides long-term voice correction; continuous power decay and head sway determine riding fatigue, announce the nearest supply point, and provide soft light reminders to rest. Scene 3: Mountain / Gravel Off-Road

[0016] The IMU uses high-frequency turbulence acceleration to distinguish between gravel, steep slopes, and elevation differences; it automatically switches to a large rolling resistance coefficient to correct power calculation errors; it integrates pitch and roll angles to predict steep slope loss of control and provides early audible and visual warnings; and it triggers a vehicle-wide emergency flashing distress signal when the vehicle tilts. Scene 4: Long-distance, multi-day trek

[0017] The pure IMU gradient accumulates the total mileage of the entire ascent / descent, and there is no loss of GPS data even when it is interrupted; the refresh rate of the helmet peripherals is dynamically reduced based on the real-time riding power, and priority is given to ensuring power supply for lights and core sensors; the APP pre-stores offline route books, and relies on the IMU gradient to match the road section offline, and announces steep slopes in advance and predicts the climbing power load. Scenario 5: Multi-person team cycling with 500-meter network setup

[0018] The team shares power, gradient, and speed in real time; the team leader checks the load of the entire team, and a voice reminder is given when a team member leaves the formation; the team leader's braking / steering star flashes in sync with the team's lights; the team's taillights automatically brighten on steep slopes; bone conduction enables real-time intercom within the team, and automatically broadcasts the team's lowest load and steep slope warnings.

[0019] Universal for all scenarios: Automatically corrects rolling resistance and wind resistance coefficient in low temperature rain and snow, improving the accuracy of slope and power calculation.

[0020] 8. Low-speed environment warning and multi-terminal alarm strategy The default light thresholds are 50 lux for light sensitivity and 5 km / h for low speed, and both thresholds can be customized; rain-related reflectivity and noise reduction shield against false triggering at low speeds; alarms cover five scenarios: sensor malfunction, communication interruption, low battery, tipping, and mechanical failure, with synchronized hardware yellow light, APP pop-up, SMS, wearable device, and bone conduction five-fold reminders.

[0021] 9. BLE5.3 + StarShine SLE Dual-Mode Encryption Binding and Network Communication The AES encryption activation process is complete; it features three levels of account permissions and dual unbinding authorization; BLE offline caching of power and gradient time-series data with customizable caching duration; StarShine SLE 500-meter networking enables team data sharing, intra-team intercom, and distress message relay forwarding; and it supports exporting cloud-based trajectory and segmented power reports.

[0022] 10. Unified management of power modules and batteries for multiple components. The percentage battery level display has been removed; the default battery life statistics period is 5 days, with a quick switch between 3 and 7 days, and support for customizing any number of days; historical power consumption of the entire device for the selected period is retrieved, and the remaining available lighting time is dynamically calculated based on real-time power, communication load, and temperature; the cumulative usage hours of the helmet, handlebars, and each light group are calculated separately, and summarized on the central control and APP independent battery pages; all three default power management values ​​are customizable; the refresh rate of peripherals is dynamically adjusted based on real-time riding power to achieve energy saving; low-temperature power consumption compensation ensures a default battery life of over 15 hours under standard operating conditions; the remaining charging time is displayed synchronously with the charging status.

[0023] 11. Dual vehicle model adaptation switching logic Human-powered bicycles: Only Hall effect sensors and IMUs are used to identify deceleration, and the software estimates the power. Electric bicycles: Accurate identification of coasting deceleration when connected to the ignition signal line. All parameters such as gradient, power, duration, and battery level have default values ​​plus custom settings. Offline gradient statistics based on IMU and matching of off-route manuals are performed.

[0024] 12. Overall system protection and expansion hardware interfaces Magnetic / contact dual charging, IPX6 waterproof seal for the whole device, and sealed anti-vibration sensor module; five types of hardware protection for the charging circuit; the central control unit has a reserved dedicated ADC expansion interface for torque sensor, and high-end configurations can achieve high-precision pedaling power measurement. 3. Beneficial effects

[0025] 1. The six-axis IMU independently calculates the slope, completely eliminating the dependence on GPS. The function of tunnels and cycling trainers is uninterrupted, and the slope is used as the core input in all scenarios. 2. Dual-mode cycling power system: the standard version has no hardware cost, professional high-configuration and high-precision calculation, automatic FTP and TSS training load quantification, covering both mass and professional cycling. 3. Dual deceleration judgment based on instantaneous deceleration and 20-second rolling speed perfectly covers both braking and coasting without braking. Automatic threshold correction on slopes significantly reduces false triggering. 4. The StarShine SLE enables multi-device networking over a distance of 500 meters, achieving team data sharing, synchronized lighting across the entire team, intra-team intercom, and distress relay, overcoming the limitations of traditional Bluetooth point-to-point communication. 5. A complete set of adaptive strategies for five major riding scenarios, integrating rear radar warning, collision distress signal, vibration anti-theft, chain / wheelset / bearing mechanical fault diagnosis, riding posture correction, fatigue reminder, and off-route climbing statistics; 6. All time, speed, deceleration, light, power, resistance, and slope thresholds are unified with "default preferred values ​​+ user-defined values" to maximize the scope of patent protection and avoid competitors' simple parameter modification to bypass protection; 7. Based on multi-day historical power consumption calculations, the actual battery life is calculated, abandoning abstract percentages. The central control and APP dedicated pages summarize the cumulative working time of all peripheral batteries, and the low battery levels can be customized. 8. AES encryption three-level account binding, hardware can operate completely independently without the mobile phone, offline time-series caching, cloud data export, and device anti-code theft; 9. Based on real-time riding power, the peripheral power consumption is dynamically adjusted, and with low temperature power consumption compensation, IPX6 whole machine waterproofing, dual charging and multiple power protection, it has strong adaptability to outdoor rain, snow and bumpy conditions. 10. Hall effect sensor + multiple IMU redundancy and fault tolerance: automatic switching to backup circuit in case of any sensor failure, ensuring uninterrupted lighting and safety warnings; compatible with both human-powered and electric bicycles, providing full coverage with a single system. Attached Figure Description

[0026] Figure 1 is a schematic diagram of the overall hardware architecture of the system of the present invention. The corresponding numbers are: 1 - Handlebar integrated sensor module, 2 - Smart helmet module, 3 - STM32 central control host, 4 - Distributed six-axis IMU unit, 5 - Millimeter-wave radar, 6 - Turn signal group, 7 - Brake light group, 8 - Deceleration warning light group, 9 - Warning light group, 10 - BLE5.3 communication unit, 11 - Star Flash SLE communication unit, 12 - Three-terminal APP, 13 - Rechargeable power supply module, 14 - Waterproof charging interface; solid lines represent power supply, dashed lines represent signals. Figure 2 shows the logic block diagram of a six-axis IMU without GPS for slope calculation and dual-mode power operation. The numbers in the figure correspond to the following modules: 1-IMU attitude acquisition unit, 2-slope smoothing compensation submodule, 3-GPS fusion correction branch, 4-software power estimation calculation unit, 5-torque sensor measured power branch, 6-power data output interface; the arrows indicate the direction of attitude, slope, and power data transmission.

[0027] Figure 3 is a block diagram of smooth deceleration dual judgment and three-level braking and lighting priority control. The numbers in the figure correspond to the following modules: 1 - instantaneous deceleration judgment branch, 2 - rolling average vehicle speed calculation unit, 3 - vehicle speed reduction comparison sub-module, 4 - three-level braking priority judgment main module, 5 - multi-mode lighting output branch; the arrows indicate the flow of vehicle speed, deceleration and lighting control signals.

[0028] Figure 4 shows the logic block diagram of adaptive lighting linkage for five cycling scenarios. The numbers in the figure correspond to the following modules: 1 - Commute scenario template unit, 2 - Road training scenario template unit, 3 - Mountain off-road scenario template unit, 4 - Long-distance crossing scenario template unit, 5 - Team cycling scenario template unit, 6 - Millimeter-wave radar early warning branch, 7 - IMU road condition recognition submodule, 8 - Star flash team lighting synchronization branch; the arrows indicate the flow of road condition, radar, and lighting linkage control signals.

[0029] Figure 5 is a block diagram of the power supply and multi-component battery management branch. The numbers in the figure correspond to the following modules: 1-SOC real-time detection unit, 2-custom cycle historical power consumption calculation module, 3-multi-component battery duration summary unit, 4-three-level low power level control sub-module, 5-low temperature power consumption compensation unit; the arrows indicate the flow of battery data, range calculation, and power consumption control signals.

[0030] Figure 6 is a block diagram of the StarShine SLE multi-device networking communication logic. The numbers in the figure correspond to the following modules: 1 - Squad data sharing unit, 2 - Intra-squad bone conduction intercom branch, 3 - SOS message relay forwarding submodule, 4 - AES encrypted binding link; the arrows indicate the flow of networking interaction, voice, alarm, and encrypted communication signals. Detailed Implementation

[0031] Example 1: Standard Commuter Version (without optional torque sensor) The hardware includes handlebar module, millimeter-wave radar bone conduction helmet, STM32 central control, complete six-axis IMU, lighting assembly, dual-mode communication, and waterproof power supply; 1. The six-axis IMU independently calculates the slope, operates normally in tunnels without GPS, and the APP allows users to customize the weight of people and vehicles and wind resistance parameters to estimate power, outputting instantaneous power and calories. 2. The deceleration logic defaults to a 20-second window and a 5% reduction. The deceleration light automatically illuminates when coasting without braking, and the deceleration threshold is automatically corrected on slopes. 3. Equipped with a full suite of commuting functions: automatic dimming for potholes, full illumination in tunnels, millimeter-wave rear-view tiered warning, tipping buffer distress message, static vibration anti-theft, and mechanical component maintenance reminders; 4. Team formation for the 500-meter Star Flash event, with synchronized lighting by the team leader and simple intercom within the team; 5. The battery has a default 5-day statistics cycle, supports switching between 3 and 7 days, has no percentage power display, and displays the cumulative usage time of accessories uniformly in the central control / APP. Three low power thresholds can be customized. 6. The default values ​​are 50 lux of light, 5 km / h low speed, fall angle, and shutdown delay. These settings are fixed in the APP and can be customized. The phone can be disconnected and the phone can run all functions independently.

[0032] Example 2: Professional Highway Training High-End Version (with Torque Sensor Added) Based on the hardware of Example 1, a center axis torque sensor is added. The central control ADC channel collects torque T and cadence ω to calculate high-precision measured power. The system automatically calculates FTP at a constant speed for 20 minutes, divides the power range into five levels, and calculates TSS training load. Dual IMU fusion identifies poor riding posture and provides voice correction. Power attenuation and head sway determine fatigue and provide replenishment. Mountain mode automatically switches to off-road rolling resistance coefficient to correct power error. Long-distance riding relies on IMU to accumulate climb and match off-road routes to predict climbing load. Dynamic power consumption scheduling extends battery life. The remaining lighting, alarm, and battery logic are completely consistent with Example 1.

[0033] Example 3: Electric Bicycle Adaptation Mode One-click switching between electric vehicle models; precise recognition of coasting deceleration when the accelerator is released via the ignition signal line; all parameters including gradient, power, lights, and battery retain default values ​​and support customization; offline gradient statistics for the entire climb based on a six-axis IMU; offline broadcast of steep slopes from the route manual.

Claims

1. A dual-model adaptive intelligent riding lighting control system integrating turn signals, three-level brake lights, gradual deceleration lights, low-speed nighttime visibility indicators, and multi-component battery graded warnings, characterized in that: The system includes a handlebar integrated sensor module, a smart riding helmet module, an STM32 embedded central control unit, a distributed six-axis IMU unit, wheel speed Hall sensors, cadence Hall sensors, a photosensor, millimeter-wave radar, turn signal group, brake light group, deceleration warning light group, general warning light group, bone conduction voice module, BLE5.3 + StarSpark SLE dual-mode communication module, Android / iOS / HarmonyOS three-terminal APP, rechargeable power supply module, and waterproof charging interface; the central control unit has a reserved ADC expansion acquisition interface for the bottom bracket / crank torque sensor, which is a professional high-end optional hardware and does not need to be added to the standard version; The STM32 embedded central control unit locally stores all operating parameters, timing cache data, IMU slope calculation formulas, and dual-mode cycling power calculation formulas, allowing the hardware to operate completely independently without the app. The BLE5.3 + StarShine SLE dual-mode communication module enables bidirectional encrypted interconnection between the app, handlebar module, helmet module, and central control unit. StarShine SLE supports low-latency multi-device concurrent networking up to 500 meters, used for transmitting IMU attitude, speed, and power data, synchronizing team lighting, reporting fall alarms, sharing team movement data, and forwarding SOS messages. BLE5.3 is used for mobile phone pairing, parameter distribution, and device status monitoring. Lighting, sensor, power, and slope threshold parameters distributed by the app are permanently stored in the central control unit. The system supports one-click switching between human-powered and electric bicycles; it features a six-axis IMU module for GPS-free slope calculation, a three-axis accelerometer to extract gravity components for coarse pitch angle calculation to obtain the basic slope, and a three-axis gyroscope to compensate for bumps and vibrations to smooth the slope value. It continuously outputs slope values ​​in GPS-free scenarios such as tunnels, underground garages, and indoor cycling trainers. When GPS signal is normal, it integrates altitude correction to mitigate long-term temperature drift errors of the IMU. The central control unit has a built-in dual-mode riding power calculation unit. The standard version estimates power using Hall effect speed and IMU slope via a three-stage resistance formula software. The high-end version integrates a torque sensor to collect torque and cadence conversion for high-precision power measurement. The software estimation includes rolling resistance, climbing gravity, and air resistance power superposition calculation, automatically outputting instantaneous / average / peak power, work done, calories, FTP threshold, five power ranges, and TSS training load. It employs Hall effect + multi-node six-axis IMU. Weighted fusion verification of turn signal triggering, built-in bump filtering and slope attitude compensation algorithms, automatic switching of remaining sensor branches in case of single-path sensor failure, and uninterrupted lighting function. The system has three levels of light priority, from highest to lowest: fall emergency flashing light, brake warning light, and turn signal light; it includes three linkage modes: brake priority, turn signal and brake coexistence, and fall highest priority; it is configured with two sets of smooth deceleration judgment logic running simultaneously: the first set is IMU instantaneous deceleration classification judgment, and the second set is rolling average vehicle speed judgment. The rolling statistics duration and vehicle speed reduction threshold can be customized; when the average vehicle speed reduction exceeds the threshold at cruising speed, the smooth deceleration warning light will automatically illuminate without mechanical braking. The system utilizes IMU gradient, real-time power, millimeter-wave radar, and vibration sensing to achieve adaptive lighting linkage across five cycling scenarios: urban commuting, professional road training, mountain gravel off-roading, long-distance multi-day treks, and group riding. IMU vibration combined with gradient detection identifies paved roads, potholes, and steep gravel slopes, dynamically adjusting light brightness and flashing frequency. In off-road conditions, it automatically adjusts the rolling resistance coefficient to correct power errors. In tunnel environments, it automatically turns on all lights. Millimeter-wave radar integrates speed and gradient to provide tiered helmet vibration, bone conduction voice, and light warnings for vehicles approaching from behind. During group riding, the system synchronizes the leader's braking and turning signals with synchronized flashing of all team lights, and automatically brightens all team taillights on steep slopes. The system includes five built-in scenario parameter templates, each with default values ​​for gradient, resistance, power, lighting, duration, and battery thresholds, and supports user customization. In the road training scenario, dual IMU fusion identifies poor riding posture and provides voice correction; continuous power decay and significant head movement indicate riding fatigue and announce aid stations. In the long-distance scenario, pure IMU gradient data accumulates ascent / descent. Downhill mileage, combined with the offline prediction of climbing power load using the APP's off-line guide; The system integrates multiple safety judgment strategies: photosensitive + vehicle speed dual threshold control for alternating low-speed parking lights at night, with customizable light threshold and low-speed threshold; In rainy weather, the system activates photosensitive reflective noise reduction logic to avoid false triggering of low-speed width indication; a six-axis IMU impact detection identifies vehicle tipping, locally caches motion timing data, automatically sends an SOS SMS when GPS is available, and caches the message for reconnection and resend when GPS is unavailable; when the vehicle is stationary, the IMU performs low-power vibration monitoring, triggering an audible and visual anti-theft alarm upon abnormal shaking; the IMU collects vibration spectrum data to identify abnormal wear on chains, wheelsets, and bearings, and pushes vehicle maintenance reminders; in low-temperature rain and snow environments, it automatically corrects rolling resistance and wind resistance coefficients to improve the accuracy of slope and power calculations; the tipping angle and duration for tipping determination are default parameters and can be customized; five types of alarms—sensor failure, communication interruption, low battery, vehicle tipping, and mechanical failure—are simultaneously pushed through hardware yellow lights, APP pop-ups, bone conduction voice, mobile SMS, and multiple reminders from smart wearables; The dual-mode communication module is equipped with AES encrypted device binding links, and can set three levels of account permissions: administrator, sub-account, and visitor. Changing the binding or adding a device requires double unbinding authorization from the original administrator. The BLE link offline caches power and gradient time-series data, with customizable caching duration. After reconnection, it synchronizes to the cloud and exports cycling tracks and segmented power reports. The StarFlash networking supports real-time bone conduction intercom within the team and automatically broadcasts the team's load and steep slope section prompts. The rechargeable power module integrates real-time lithium battery SOC detection, eliminating the pure percentage power display; the default battery life statistics period is 5 days, with quick switching between 3 days and 7 days, and also supports customization of any duration; the central control retrieves the historical power consumption data of the whole machine within the period, and dynamically calculates the remaining usable lighting time by combining real-time power, communication load, and ambient temperature; It independently collects power consumption data from the helmet, handlebars, and each light assembly, and summarizes and displays the cumulative working time of each accessory on the central control unit and the dedicated battery page in the APP; the system presets three low-battery protection thresholds, and the threshold values ​​can be customized; it dynamically adjusts the refresh rate of peripherals based on real-time riding power to achieve energy saving, and prioritizes power supply to lights and core sensors during high-load climbing and night riding; it is equipped with a low-temperature power consumption compensation algorithm, and the standard working condition single full charge can last for no less than 15 hours; the waterproof charging interface is compatible with both magnetic and contact charging, the whole device is IPX6 fully sealed waterproof, and the charging circuit integrates five types of hardware protection circuits for overvoltage, overcurrent, overtemperature, short circuit, and reverse connection.

2. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The three-level brake light priority is divided from high to low as follows: mechanical lever brake light, emergency braking strobe light, and gradual deceleration light; the gradual deceleration light is directly disabled after the mechanical brake is triggered; the instantaneous deceleration judgment is divided into three ranges: gradual deceleration 0.3~1.2m / s², moderate braking 1.2~2.5m / s², and emergency braking >2.5m / s², and each threshold value can be customized; the IMU identifies the vehicle pitch angle in real time and dynamically corrects the slope deceleration threshold, adjusting it uphill and downhill to eliminate false triggering of deceleration caused by slope gravity; emergency braking uses a dual condition verification of sudden speed drop + high deceleration; The rolling average vehicle speed determination uses a default 20-second statistical window and a 5% cruise speed reduction threshold, both of which can be customized. Human-powered bicycles rely on Hall effect speed recognition to identify coasting, while electric bicycles have an additional throttle signal line for accurate recognition of coasting deceleration when the throttle is released.

3. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The turn signal system collects the steering angle and duration from the handlebar Hall sensor, and combines this with multiple conditions such as vehicle speed, vehicle tilt angle, and vibration from multi-node IMUs for joint judgment. The turn signal delay time when the handlebar returns to center is a customizable default parameter. In the event of a fault in the Hall sensor or any single IMU, the system removes the abnormal data and continues to control the turn signal using the remaining sensor units. The gradient value calculated independently by the six-axis IMU without GPS serves as the core input parameter for adaptive dimming, riding power calculation, road condition recognition, and total elevation gain statistics.

4. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The emergency flashing mode takes priority over brakes and turn signals when a fall occurs. When the IMU detects that the vehicle's tilt angle exceeds a set threshold and continues for a preset duration, all lights on the vehicle will flash at a high frequency. The tilt angle and judgment duration can be customized. The emergency mode will automatically deactivate once the vehicle returns to its upright position. When riding in a group, the Star Flash SLE mode synchronously transmits the leader's turn and brake commands, and the corresponding lights of the entire group will light up synchronously. The system dynamically adjusts the brightness and flashing frequency of the lights based on real-time riding power, road gradient, and road condition.

5. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The locally adjustable parameters for the smooth deceleration lights include cruise speed window, speed anti-shake threshold, deceleration confirmation delay, rolling statistics duration, speed reduction percentage, light brightness for various road conditions, and flashing frequency. All parameters are equipped with system default values. The light output supports two modes: slow flash and constant light. The taillight, helmet warning light, and pedal lights can be individually controlled and grouped for activation and deactivation. The three-terminal APP has built-in parameter templates for five scenarios: commuting, highway, mountain, long-distance, and team riding. The gradient coefficient, rolling / wind resistance, power threshold, light duration, and low battery level in the templates can all be modified with one click and permanently saved to the central control unit. In the software's power estimation formula, the total weight of the rider and vehicle, tire rolling resistance coefficient, frontal area, and transmission efficiency are customizable default parameters. With the optional torque sensor, high-precision pedaling power is calculated using the formula P_measured = T・ω, automatically calculating the FTP threshold, dividing the power range into five levels, and calculating the TSS training load for a single ride.

6. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The low-speed parking light is triggered at night when the photosensitive light intensity is below 50 lux and the real-time vehicle speed is below 5 km / h. The light intensity threshold and low-speed vehicle speed threshold can be customized. When rainwater reflection interferes with the photosensitive sampling, the sensor noise reduction and locking mechanism is activated to prevent the low-speed parking light from being triggered erroneously. Millimeter-wave radar integrates vehicle speed and slope to identify the distance to vehicles approaching from behind, triggering a triple warning system of helmet vibration, bone conduction voice, and taillight warning; four types of alarms—equipment failure, low battery, vehicle tipping, and chain / wheelset / bearing mechanical failure—are simultaneously pushed out via hardware yellow light, APP pop-up, mobile phone SMS, and smart wearable device reminders.

7. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The BLE5.3+ StarFlash SLE AES encryption binding process is as follows: Upon powering on the hardware, a Bluetooth data packet with a verification key is broadcast, and an activation QR code containing a unique serial number is displayed; the APP scans the code to extract the serial number and communication key, completes a two-way handshake verification with the central control unit, and after successful verification, the device serial number is permanently bound to the administrator account; Each hardware device is bound to only one administrator. Adding a sub-device or changing the bound account requires dual authorization from the original administrator. The BLE link offline caches cycling, power, and gradient time-series data with customizable caching duration. After reconnection, it automatically synchronizes to the APP cloud. The StarShine SLE 500-meter network enables team power, gradient, and speed data sharing, intra-team bone conduction intercom, and distress message relay forwarding. The cloud supports storing cycling tracks and exporting segmented power reports.

8. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The power module lights use intermittent flashing low-power drive, and the continuous warning time under standard working conditions is ≥15 hours when fully charged; the system does not display a simple battery percentage, and the battery life calculation cycle is 5 days by default, and supports quick selection of 3 days and 7 days or manual customization of the number of statistical days; the central control retrieves the historical power consumption of the whole machine in the selected period, and dynamically calculates the remaining light usage time by combining the number of lights turned on, dual-mode communication load, ambient temperature, and real-time riding power; The system collects independent power consumption data for the helmet, handlebars, and each light assembly, and visualizes and summarizes the cumulative usage hours of each accessory on the central control display and the battery management page of the three-terminal app. The system has three preset power control logics: >30% power, full operation; 20%~30% power, push battery life reminder; 10%~20% power, turn off auxiliary lights; <10% power, only the brake lights are kept on. The threshold values ​​for the three levels can be customized. It is equipped with a low-temperature power consumption compensation algorithm to correct the estimated battery life in low-temperature environments and reduce the duty cycle of the lights to extend the battery life. The system dynamically allocates the operating refresh rate of each peripheral based on real-time riding power, and the charging status is synchronized with the hardware indicator lights and the remaining charging time is displayed on the app.

9. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The magnetic and contact charging ports are integrated into the whole unit, which adopts an IPX6 fully sealed waterproof structure. The six-axis IMU, photosensitive sensor, and millimeter-wave radar are equipped with a sealed anti-shake structure to adapt to bumpy outdoor riding. The charging circuit integrates five types of hardware protection circuits for overvoltage, overcurrent, overtemperature, short circuit, and reverse connection, so there is no risk of leakage or short circuit when riding in the rain. The central control host has a reserved ADC expansion acquisition interface for the bottom bracket / crank torque sensor. After connecting the torque sensor, it can switch to a high-precision power measurement mode.

10. The dual-model adaptive intelligent riding lighting control system according to claim 1, characterized in that: The system's vehicle mode switching logic is as follows: In human-powered bicycle mode, the system relies solely on Hall sensors and a six-axis IMU to identify deceleration and steering actions, and uses software algorithms to estimate riding power throughout the process; in electric bicycle mode, the system connects to the throttle signal line to accurately identify coasting deceleration behavior when the throttle is released; in both vehicle modes, all gradient thresholds, power-drag coefficients, light durations, and power management values ​​are equipped with system default values, which can be customized and distributed via the APP and permanently stored locally; Based on the six-axis IMU's offline slope data without GPS, the system can statistically analyze the entire ascent and descent mileage, match it with the APP's pre-stored offline route guide, and provide offline broadcasts of steep slopes ahead and predict the climbing power load.

Images