Motorcycle riding behavior analysis system and riding behavior analysis program

Abstract

Conventional motorcycle riding analysis required numerous additional physical sensors, resulting in complex systems and geometrically constrained mounting positions. Furthermore, evaluation using fixed parameters was difficult in environments where road surface conditions constantly changed, making it impossible to quantitatively calculate the dynamic state of the suspension or areas for improvement in rider operation. [Solution] The system includes a characteristic estimation means that calculates the deviation between the theoretical value and the measured value of the dynamic equilibrium state using vehicle-specific geometric and physical parameters, based on driving data acquired from at least one inertial measurement unit (IMU) and a global navigation system (GNSS). This estimates the internal state, including the dynamic sag amount and damping characteristics of the suspension. Furthermore, it normalizes time-series data of multiple laps with the driving distance as the reference axis and automatically outputs vehicle setting guidelines and riding improvement guidelines using a common algorithm, from a minimal configuration with a single unit to a distributed configuration with multiple units.

Description

[Technical Field]

[0001] The present invention relates to a running behavior analysis system that applies a vehicle-specific geometric and mechanical physical model to running data acquired from an inertial measurement unit (IMU) and a global navigation satellite system (GNSS) mounted on a straddle-type two-wheeled vehicle (hereinafter referred to as "two-wheeled vehicle") including road racing motorcycles, dirt track racing vehicles, motocross vehicles, enduro vehicles, motard vehicles and the like, to inversely estimate the dynamic internal state of portions where sensors are not directly arranged, and automatically generate quantitative evaluation of suspension characteristics, tire conditions and road surface properties as well as improvement guidelines for rider operations. [Background Art]

[0002] Conventionally, mounting of multi-point measurement devices such as suspension stroke sensors, wheel speed sensors, and steering torque sensors has been required to accurately grasp the suspension state of a two-wheeled vehicle. However, these devices are complicated to install, have problems in reliability in harsh driving environments (vibration, temperature, mud, sand, water, etc.), and have an essential problem that the mounting of the sensors itself changes the geometric characteristics of the vehicle. Especially in off-road competitions such as motocross (MX), supercross (SX), and enduro, road surface properties change significantly every lap, making it difficult to apply evaluation methods based on fixed thresholds.

[0003] Existing GPS data loggers (see Patent Document 1) focus on recording and visualizing position, speed, and running trajectory, and do not have functions for frequency analysis-based estimation of suspension damping characteristics or quantitative evaluation of rider operations. In addition, normalization calculation for comparing a plurality of laps with different lap times in the same distance coordinate system is not disclosed.

[0004] Estimation methods using the equation of motion of the vehicle body (see Patent Documents 2 and 3) are known, but these methods are based on multi-point sensor information or only estimate attitude angles, and do not include functions for estimating suspension damping characteristics through frequency analysis, dynamically estimating road surface properties, or quantitatively evaluating rider operations. [Prior Art Documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 5344142 (DigiSpice Co., Ltd.: Mobile Object Driving Analysis System) [Patent Document 2] Japanese Patent Publication No. 6478263 (Yamaha Motor Co., Ltd.: Motorcycle Behavior Analysis Device) [Patent Document 3] Japanese Patent Publication No. 2020-134148 (Honda R&D Co., Ltd.: Vehicle attitude estimation device) [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] Conventional driving analysis systems are limited to recording positional information using GNSS and simply displaying attitude angles, making it difficult to calculate internal conditions directly related to the vehicle's physical setup, such as suspension damping characteristics, dynamic sag, tire condition, and road surface characteristics. Furthermore, understanding these requires attaching numerous physical sensors, which presents the challenge of geometric constraints on sensor mounting positions. Moreover, in off-road competitions where road surface characteristics change with each lap, there is a fundamental problem: evaluation methods based on fixed parameters do not function effectively.

[0007] The present invention aims to provide a common analysis algorithm, ranging from a minimal configuration with a single unit to a distributed measurement configuration with multiple units, that couples the vehicle's geometric and physical model with the equations of motion, thereby presenting suspension conditions, road surface conditions, and riding improvement guidelines, including areas without sensors, by coupling these with the vehicle's geometric and physical model and equations of motion, in a configuration that includes at least one inertial measurement unit. [Means for solving the problem]

[0008] The present invention provides a characteristic estimation means for estimating the dynamic internal state of a vehicle, including at least one of the suspension characteristics, tire condition, and road surface characteristics of a motorcycle, by calculating the deviation between a theoretical value of the dynamic equilibrium state calculated from the GNSS data and a measured value from the IMU, using vehicle parameters that define the vehicle's unique geometric and inertial characteristics based on driving data acquired from at least one inertial measurement unit (IMU) and a global navigation system (GNSS) mounted on the motorcycle.

[0009] The characteristic estimation means calculates the dynamic sag amount based on the correlation between the theoretical vertical acceleration derived from the bank angle and vehicle speed during cornering and the measured vertical acceleration, or estimates the damping characteristics based on the power spectral density analysis of IMU data in a frequency window dynamically set according to the wheel speed and vehicle speed. It also includes means for normalizing the driving data to a common distance axis based on the GNSS data, generating the theoretical fastest lap from the data of multiple laps, and extracting items for improvement in riding operation. Furthermore, it includes a configuration in which measurement units are distributed to multiple parts of the vehicle body, and the internal state is calculated based on the phase difference and amplitude ratio between each unit. [Effects of the Invention]

[0010] According to the present invention, from a minimum configuration of at least one inertial measurement unit to a configuration of multiple units distributed across multiple locations (such as a distributed 4IMU configuration), a common geometric physical model and equations of motion can be coupled to automatically calculate the internal state of the suspension, road surface conditions, and indicators for improving rider operation, which previously required additional physical sensors such as stroke sensors, across all categories from road racing to off-road competitions.

[0011] Furthermore, distance normalization calculations enable objective comparisons independent of lap time differences, and the consistent algorithm provides professional-level setup guidelines, from simple single-unit measurements to high-precision distributed measurements using multiple units. In addition, even in transient environments where road surface conditions change, the internal state can be inversely calculated from the deviation of the dynamic equilibrium state, enabling analysis that responds immediately to the environment. [Brief explanation of the drawing]

[0012] [Figure 1] Overall configuration and functional block diagram of the driving analysis system. [Figure 2] Geometric coupling model and internal state calculation diagram (Equation (3) β_mech, Equation (5) a_proj, Equation (6) Fz_coupled). [Figure 3] Damping characteristics and road surface condition estimation diagram based on PSD analysis (Equation (4) E_band). [Figure 4] A diagram showing the calculation of dynamic sag based on the deviation between theoretical and measured vertical acceleration (Equation (1) Az_theory, Equation (2) Δh_dyn). [Figure 5] Distance-normalized comparison chart of multiple laps (generating theoretical fastest lap data). (Legend) 5a: Best lap (solid line), 5b: 2nd lap (dashed line), 5c: ΔSpeed ​​B-2nd, 5d: ΔAy B-2nd, T1 to T6: Corner apex (speed minimum) [Figure 6] Quantitative evaluation diagram (Ax, Ay, Gz) and improvement guideline diagram of lidar operation using time-series waveforms. [Figure 7] Coherence analysis diagram between distributed units (Equation (7)). [Modes for carrying out the invention]

[0013] [Definition of symbols and notations] The main symbols and terms used in this specification are defined as follows in Table 1. [Table 1]

[0014] [Technical Features of the Present Invention] The traveling analysis system of the present invention is essentially characterized by inversely calculating the internal state of a portion not equipped with sensors by coupling traveling data obtained from at least one inertial measurement unit (hereinafter referred to as "IMU") and a satellite positioning system (hereinafter referred to as "GNSS") with a vehicle-specific geometric physical model. Specifically, it has remarkable inventive step in the point that the following three elements are integrated. First, a distance normalization calculation using the GNSS integrated distance as a reference enables comparison of data from a plurality of laps with different lap times in the same coordinate system. Second, based on power spectral density (PSD) analysis of IMU data, the damping characteristics of a suspension and road surface properties can be estimated in a non-contact manner. Third, based on a deviation between a theoretical value and an actually measured value in a dynamic equilibrium state, internal states such as a dynamic sag amount and a ground contact load can be inversely estimated.

[0015] [System Configuration and Operation Specifications] The minimum configuration and recommended specifications of the traveling analysis system according to an embodiment of the present invention are shown in Table 2 below. Note that the measurement unit 100 may have a single configuration, or may be configured by a plurality of units dispersedly arranged at respective parts of a vehicle body.

Table 2

[0016] [Details of Calculation Algorithm] The analysis calculation unit 20 calculates a theoretical vertical acceleration Az_theory (G) by the following formula (1) based on a vehicle speed v (m / s) and a turning radius R (m) obtained from the GNSS 2, and a bank angle φ (rad) estimated from the IMU 1.

Math

[0017] [Estimation of Dynamic Sag Amount] The internal state estimation means 23 calculates the deviation ΔAz between the measured vertical acceleration and the theoretical vertical acceleration, and estimates the dynamic sag amount Δh_dyn(m) using the following equation (2) with respect to Iy, M, and Lf read from the vehicle parameter holding means 24.

number

[0018] [Calculation of mechanical vehicle slip angle] The preprocessor 10 calculates the mechanical body slip angle β_mech(rad) using the following equation (3), based on the phase difference between the time integral value of the deflection angular velocity ψ·_IMU obtained from IMU1 and the direction of travel angle ψ_path obtained from GNSS2.

number

[0019] [Estimation of damping characteristics] The analysis unit 20 calculates the energy intensity E_band(G^2) of a specific frequency band using the following equation (4) for the power spectral density S(f) of the IMU1 data, and determines the damping characteristics of the suspension.

number

[0020] [Estimation of dynamic ground load] Furthermore, the internal state estimation means 23 calculates the projected acceleration a_proj(G) in the direction of the road surface normal using the following equation (5), and estimates the dynamic ground contact load Fz_coupled(N) using equation (6), taking into account the geometric coupling term.

number

number

[0021] [Evaluation in distributed configurations] In a configuration in which measurement units are distributed across multiple locations, the characteristic estimation means 23 verifies the suspension stroke state and the dynamic rigidity of the vehicle body based on the coherence γ^2(f) calculated by the following equation (7) (see Figure 7).

number

[0022] [Examples] As an embodiment demonstrating the effectiveness of the present invention, an example of applying this system to a two-wheeled racing vehicle will be described. This system pre-stores vehicle-specific parameters such as the wheelbase, center of gravity, suspension stroke length, and expected static sag amount of the target vehicle in the parameter holding means 24. During driving, the system couples these parameters with measurement data, The time-series analysis shown in Figure 6 identified a section (a gap in operation) where the braking deceleration Ax remained at a predetermined threshold, and the rise of the yaw rate Gz, which indicates the lean-in speed, was delayed relative to the distance axis. This quantitatively identified a failure in load transfer caused not by a suspension setting defect, but by the rider's operating characteristics (insufficient turn-in overlap). Furthermore, analysis in Figure 7 revealed that when the coherence maintenance rate in the stable corridor during cornering (0.5 to 7.5 Hz) fell below the target value, and the wobble component around 9.0 Hz increased, adjustment guidelines for the rear damping (such as increasing rebound damping) were output. These automatically execute the following judgment logic.

[0023] [Evaluation logic for dynamic sag amount] The system calculates the dynamic sag amount Δh_dyn from equation (2) based on the difference between the theoretical vertical acceleration calculated by equation (1) and the measured vertical acceleration during the braking phase of corner entry. If the calculated dynamic sag amount falls below the "target dive amount (predetermined range) of the vehicle" stored in the parameter holding means 24, the system determines that the front fork has not compressed sufficiently. Furthermore, the system refers to the braking acceleration (deceleration) calculated simultaneously, and if the deceleration does not reach the "predetermined braking threshold," it determines that the main cause of insufficient front wheel load is insufficient braking input by the rider, rather than a defect in the physical settings of the suspension (spring rate, etc.), and outputs improvement guidelines encouraging increased braking.

[0024] [Evaluation logic for stable corridors based on coherence] In a configuration using IMUs distributed across multiple locations, the system calculates the coherence γ^2(f) between each measurement unit during extreme cornering (for example, when the rear tires are sliding). The system monitors the percentage (maintenance rate) in which the calculated coherence maintains a "predetermined reference threshold" in the low-frequency band (a predetermined frequency band corresponding to the stable corridor during extreme driving), which indicates the stability of the vehicle's behavior. If a section is detected where this maintenance rate falls below the "target rate," the system determines that there is a loss of tire contact load or that the vehicle is in an excessive sliding state.

[0025] [Optimization logic for damping characteristics] In addition to the coherence evaluation described above, the system monitors the peak value and energy intensity of the power spectral density in the high-frequency band (a predetermined frequency band corresponding to the suspension's wobble component, etc.). If the energy intensity exceeds a predetermined tolerance threshold, the system automatically outputs recommended setting changes to restore road-following ability, such as relaxing the compression damping (COMP) or strengthening the rebound damping (TEN) of the rear shock absorber.

[0026] [Quantitative evaluation logic for rider operation] The operation evaluation means 26 calculates the yaw rate and lateral acceleration during turn-in and analyzes the response time (turn-in overlap) until these reach a "target threshold dynamically set according to the vehicle category and road surface conditions." The system identifies the delay in the rise of these physical quantities during the transition from braking to turning and quantifies the deviation from the target simultaneous operation, thereby presenting the time that can be shortened (room for improvement) for each sector. [Notes on typical embodiments of the present invention] Typical embodiments of the technology disclosed herein are described below. [Note 1] A driving analysis system for analyzing the driving state of a motorcycle using driving data acquired from at least one inertial measurement unit (IMU) and a global navigation system (GNSS) mounted on the motorcycle, comprising: parameter holding means for holding vehicle parameters that define the geometric characteristics or inertial characteristics of the motorcycle; and characteristic estimation means for estimating the dynamic internal state of the vehicle, including at least one of the suspension characteristics, tire condition, and road surface characteristics of the motorcycle, by calculating the deviation between a theoretical value of the dynamic equilibrium state calculated from the GNSS data and a measured value from the IMU, using the vehicle parameters. [Appendix 2] The driving analysis system according to Appendix 1, comprising distance normalization means that projects the time-series driving data onto a distance axis and normalizes it using the travel distance calculated based on the GNSS data as a common reference axis, and the characteristic estimation means estimates the internal state using the driving data of multiple sections normalized on the distance axis. [Note 3] The driving analysis system according to Note 1 or 2, characterized in that the characteristic estimation means calculates the amount of dynamic suspension sink based on the difference between the theoretical vertical acceleration in the direction of the road surface normal derived from the bank angle and vehicle speed during turning and the vertical acceleration measured by the IMU. [Note 4] The driving analysis system according to any one of Notes 1 to 3, characterized in that the characteristic estimation means calculates the energy intensity in a frequency band dynamically set according to the vehicle speed or wheel speed based on the power spectral density analysis of the measured data of the IMU, and estimates the damping characteristics of the suspension or the road surface conditions based on said energy intensity. [Appendix 5] A driving analysis system according to any one of Appendix 1 to 4, characterized in that it comprises a calculation means for calculating a mechanical vehicle slip angle corresponding to the phase difference between the directional vector of the vehicle axis and the trajectory vector of the contact point, based on the difference between the time integral value of the deflection angular velocity obtained from the IMU and the direction of travel angle obtained from the GNSS. [Note 6] The driving analysis system according to any one of Notes 1 to 5, characterized in that the inertial measurement unit is composed of a plurality of measurement nodes distributed to a plurality of parts of the two-wheeled vehicle, and the characteristic estimation means calculates the internal state based on the phase difference and amplitude ratio of acceleration or angular velocity between the plurality of measurement nodes. [Appendix 7] A driving analysis system according to any one of Appendix 2 to 6, characterized in that it comprises an output means that outputs at least one of the following: a recommended setting value for the vehicle, a determination result of the road surface condition, and a guideline for improving the rider's operation, based on the difference between the internal state estimated by the characteristic estimation means and the driving data compared by the distance normalization means. [Appendix 8] A driving analysis program that uses driving data acquired from at least one inertial measurement unit (IMU) and a global navigation system (GNSS) mounted on a motorcycle to cause a computer to function as one of the means described in any one of Appendices 1 to 7. [Industrial applicability]

[0027] The present invention can be widely used as a motorcycle riding data analysis device, a vehicle setting support device, and a riding diagnostic program. Specifically, it can be applied to riding data loggers as aftermarket parts, analysis applications that run on mobile devices such as smartphones, or data analysis services at racetracks.

[0028] Furthermore, the estimated suspension state and road surface conditions calculated by this invention can be fed back in real time to the vehicle's electronic control unit (active suspension, traction control, power mode control, etc.) as a control logic, making it usable in the motorcycle manufacturing industry as well.

[0029] Furthermore, it has high utility in a wide variety of competition categories, including road racing, motocross, enduro, and supermoto, as well as in all industrial fields that require the visualization of the dynamic behavior of motorcycles, such as maintenance diagnostics for public road vehicles and integration into safe driving support systems. [Explanation of Symbols]

[0030] 1 IMU (Inertial Measurement Unit) 2 GNSS modules 10 Pre-processing 20 Analysis calculation section 21 Distance normalization means 22 Means for identifying the travel section 23 Internal state estimation means 24 Parameter holding means 25. Theoretical Fastest Driving Data Calculation Method 26 Operation evaluation means 27 Road surface condition estimation means 30 Output section 40 Vehicles in operation 61 Longitudinal acceleration (Ax) 62 Lateral acceleration (Ay) 63 Yaw rate (Gz) 71 Coherence Threshold 72. Stable Corridor (Low Frequency Band) 73 Abnormal vibration components 100 measurement units

Claims

1. A driving analysis system that analyzes the driving state of a motorcycle using driving data acquired from an inertial measurement unit (IMU) and a global navigation system (GNSS) mounted on the motorcycle, A parameter holding means for holding vehicle parameters that define the unique geometric and inertial characteristics of the motorcycle, including the wheelbase, center of gravity, moment of inertia, vehicle mass, and suspension stroke length of the motorcycle, A theoretical vertical acceleration calculation means that acquires positioning data from the satellite positioning system and calculates the vehicle speed and turning radius based on said positioning data, acquires acceleration and angular velocity from the inertial measurement unit and calculates the bank angle based on said acceleration and angular velocity, and calculates the theoretical value of the vertical acceleration in the direction of the road surface normal acting on the two-wheeled vehicle during a turn based on the calculated vehicle speed, turning radius and bank angle, The inertial measurement unit obtains the component of the acceleration measured in the direction of the road surface normal as the measured vertical acceleration, and the deviation calculation means compares the measured vertical acceleration with the theoretical value of the vertical acceleration calculated by the theoretical vertical acceleration calculation means, treating them as identical physical quantities that are both vertical acceleration in the direction of the road surface normal, and calculates the difference between them. A dynamic sag estimation means estimates the dynamic sag amount, which is the amount of dynamic sinking of the motorcycle's suspension, based on the deviation calculated by the deviation calculation means and the moment of inertia, vehicle mass, and center of gravity position read from the parameter holding means. A driving analysis system characterized by having the following features.

2. The distance normalization means calculates the distance traveled from the starting position by integrating the positioning data acquired from the satellite positioning system over time, and projects each data point of the travel data acquired based on time onto a position on a distance axis that uses the travel distance corresponding to the acquisition time of the data point as the reference axis, thereby aligning travel data from multiple laps with different travel times so that data at the same position on the distance axis correspond to each other. The driving analysis system according to claim 1, characterized in that the dynamic sag estimation means estimates the amount of dynamic sag using driving data from a plurality of laps that are correlated with each other on the distance axis.

3. The driving analysis system according to claim 1, further comprising damping characteristic estimation means that calculates the power spectral density of the vertical acceleration measured by the inertial measurement unit, sets a frequency band according to the vehicle speed or wheel speed, calculates the energy intensity by integrating the power spectral density in the frequency band, compares the calculated energy intensity with a predetermined allowable threshold, determines that the damping characteristics of the suspension are in a state that impairs road surface following ability when the energy intensity exceeds the allowable threshold, and outputs recommended setting changes including relaxation of compression damping force or strengthening of rebound damping force.

4. The driving analysis system according to claim 3, further comprising a road surface condition estimation means that calculates the energy intensity in a frequency band dynamically set according to the vehicle speed or wheel speed based on the power spectral density analysis of the measurement data of the inertial measurement unit, and estimates the road surface condition based on the energy intensity.

5. The driving analysis system according to claim 1, further comprising a slip angle calculation means that calculates an azimuth angle representing the direction of the vehicle axis by integrating the deflection angular velocity obtained from the inertial measurement unit over time, and calculates an azimuth angle representing the direction of travel of the trajectory of the contact point based on the positioning data obtained from the satellite positioning system, and calculates the difference between these two azimuth angles, both of which are azimuth angles, as a mechanical vehicle slip angle corresponding to the phase difference between the direction vector of the vehicle axis and the trajectory vector of the contact point.

6. The inertial measurement unit is composed of multiple measurement units distributed across multiple locations, including the unsprung and sprung portions of the motorcycle. The driving analysis system according to claim 1, comprising a coherence evaluation means that calculates coherence from the cross spectrum of the measurement signal of the measurement unit located in the unsprung portion and the measurement signal of the measurement unit located in the sprung portion, calculates the percentage of the calculated coherence that remains above a predetermined reference threshold in a predetermined frequency band corresponding to a stable corridor during limit driving as a maintenance rate, compares the maintenance rate with a predetermined target rate, and determines and outputs that driving sections in which the maintenance rate falls below the target rate are sections in which the tire contact load is lost or excessive sliding is occurring.

7. The driving analysis system according to claim 1, further comprising an operation evaluation means that compares the amount of dynamic sag estimated by the dynamic sag estimation means with a range of target dive amounts of the motorcycle stored in the parameter holding means, and compares the braking acceleration obtained from the inertial measurement unit with a predetermined braking threshold, and if the amount of dynamic sag falls below the range of target dive amounts and the braking acceleration does not reach the braking threshold, determines that insufficient braking input by the rider is the main cause of the insufficient front wheel load and outputs improvement guidelines to encourage increased braking operation, and if the amount of dynamic sag falls below the range of target dive amounts and the braking acceleration reaches the braking threshold, determines that deficiencies in the physical settings of the suspension are the main cause of the insufficient front wheel load and outputs a statement to that effect.

8. A driving analysis program that uses driving data acquired from an inertial measurement unit (IMU) and a global navigation system (GNSS) mounted on a motorcycle to cause a computer to function as one of the means described in any one of claims 1 to 7.

Images