Failure evaluation and reliability analysis method for actuator of multi-axis unmanned differential vehicle

Abstract

The invention discloses an actuator failure evaluation and reliability analysis method for a multi-axis unmanned differential vehicle, and the method comprises the steps: numbering driving actuators, generating a failure combination set according to the failure number and spatial distribution, constructing a failure fault library, and carrying out the reliability evaluation of the actuator. The failure combination of the driving actuator of the multi-axle unmanned differential vehicle can be systematically organized and uniformly managed, and the problem that in the prior art, the failure combination with remarkable spatial distribution difference of the multi-axle vehicle is difficult to cover is effectively solved. Driving system reliability analysis is performed based on a failure database, failure of a single driving actuator is taken as a basic event, failure quantity exceeding a tolerance limit is taken as an intermediate event to construct a fault tree, and system failure probabilities under a fault-tolerant control structure and a fault-tolerant control structure are calculated respectively. And comprehensive risk judgment is carried out in combination with the total severity level, and vehicle operation constraints and fault-tolerant control intervention bases corresponding to different risk levels can be output.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, and more specifically to a method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle. Background Technology

[0002] Multi-axle unmanned differential steering vehicles typically employ distributed drive actuators (such as hub motors) to achieve traction and differential steering control, creating the desired yaw motion through differences in drive output between the left and right sides. These vehicles face challenges such as long-term operation, large load variations, and complex road conditions in engineering operations and special transportation scenarios. Failure of the chassis drive actuators can easily lead to safety risks such as decreased longitudinal dynamics, poor lateral stability, and yaw instability. Furthermore, the large number and diverse spatial distribution of actuators in a multi-axle structure mean that failures often exhibit an evolutionary pattern of "single-point failure to multi-point combined failure," with the failure location and combination mode significantly impacting the vehicle's dynamic response and risk level.

[0003] In existing technologies, research on vehicle actuator failures often separates failure detection / diagnosis, failure severity assessment, and system reliability analysis. One type of method tends to focus on qualitative / semi-quantitative assessment of single operating conditions or single actuator failures, making it difficult to cover failure combinations with significant spatial distribution differences in multi-axle vehicles, such as failures on the same side, coaxial, or diagonal. Another type of method, although it can obtain the failure probability at the system level using fault trees and other means, lacks a unified mapping with vehicle dynamics severity indicators, making it difficult to form a comprehensive risk judgment and operational constraint output oriented towards fault-tolerant control intervention. Summary of the Invention

[0004] In view of this, the present invention provides a method for actuator failure assessment and reliability analysis of multi-axle unmanned differential vehicles, in order to solve the problems that the existing technology is unable to cover failure combinations with significant differences in the spatial distribution of multi-axle vehicles, and is unable to form a comprehensive risk judgment and operational constraint output for fault-tolerant control intervention.

[0005] A method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle, comprising: Step S1: Obtain vehicle dynamics data through onboard sensors, and establish a multi-axle unmanned differential steering vehicle dynamics model based on the vehicle dynamics data; Step S2: Number the drive actuators according to the arrangement relationship of the drive actuators on the left and right sides of each axle of the vehicle, generate a set of drive actuator failure combinations according to the number of failures and spatial distribution, and build a failure fault library containing each failure combination. Step S3: For each failure combination in the failure database, perform a no-failure baseline simulation and a corresponding failure simulation under preset straight-line driving conditions and preset steering conditions, respectively, to obtain the dynamic response results of the vehicle in the no-failure state and the failure state. Step S4: Based on the dynamic response results under no-failure and failed states, construct the longitudinal severity index, lateral severity index, and yaw severity index. Step S5: Based on the longitudinal severity index, lateral severity index, and yaw severity index, each actuator failure combination is mapped to a corresponding sub-severity level, and the overall severity level of each actuator failure combination is obtained through multi-index fusion, thereby realizing the severity classification and ranking of chassis actuator failure combinations. Step S6: Analyze the reliability of the drive system based on the failure database. Construct a fault tree with the failure of a single drive actuator as the basic event and the number of failures exceeding the tolerance limit as the intermediate event. Calculate the system failure probability under the fault-tolerant control structure and the fault-tolerant control structure respectively. Step S7: Make a comprehensive risk judgment based on the overall severity level and the corresponding system failure probability, and output the basis for vehicle operation constraints and / or fault-tolerant control intervention corresponding to different risk levels.

[0006] The actuator failure assessment and reliability analysis method for multi-axle unmanned differential vehicles provided by the present invention has the following beneficial effects: (1) By numbering the drive actuators and generating a set of failure combinations according to the number of failures and spatial distribution, and constructing a failure database, the present invention can systematically organize and uniformly manage the failure combinations of drive actuators of multi-axle unmanned differential vehicles, effectively solving the problem that the existing technology is difficult to cover failure combinations with significant differences in spatial distribution of multi-axle vehicles.

[0007] (2) The present invention conducts a failure-free benchmark simulation and a corresponding failure simulation under a preset straight-going condition and a preset turning condition, respectively, to obtain the dynamic response under the failure-free state and the failure state, so that the failure effect has a consistent reference.

[0008] (3) Based on dynamic response, the present invention constructs longitudinal severity index, lateral severity index and yaw severity index, maps failure combinations to sub-severity levels, and obtains the overall severity level through multi-index fusion, effectively realizing the severity classification and ranking of chassis actuator failure combinations.

[0009] (4) This invention performs drive system reliability analysis based on a failure database. It constructs a fault tree with the failure of a single drive actuator as the basic event and the number of failures exceeding the tolerance limit as the intermediate event. It calculates the system failure probability under fault-tolerant control structure and fault-tolerant control structure respectively, and makes a comprehensive risk judgment in combination with the overall severity level. It can output vehicle operation constraints and fault-tolerant control intervention basis corresponding to different risk levels, and solves the problem that the existing technology is difficult to form a comprehensive risk judgment and operation constraint output for fault-tolerant control intervention. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the actuator failure assessment and reliability analysis method for a multi-axle unmanned differential vehicle provided in an embodiment of the present invention. Figure 2 The graph shows the vehicle dynamic response parameters for the straight-ahead fault groups 1, 2, and 3 in Table 5. Figure 3 The graph shows the vehicle dynamic response parameters for straight-ahead fault groups 6, 8, and 9 in Table 5. Figure 4 The graph shows the vehicle dynamic response parameters for straight-ahead fault groups 1, 4, and 10 in Table 5. Figure 5 The graph shows the vehicle dynamic response parameters for fault groups 5, 6, and 7 in Table 5. Figure 6 The graph shows the vehicle dynamic response parameters for fault groups 11 and 12 in Table 5. Figure 7 The graph shows the vehicle dynamic response parameters for the turning fault groups 1, 2, and 3 in Table 5. Figure 8 The graph shows the vehicle dynamic response parameters for the turning fault groups numbered 1, 4, and 10 in Table 5. Figure 9 This is a schematic diagram of a fault tree for a fault-tolerant control system. Detailed Implementation

[0011] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain embodiments of the present invention, and should not be construed as limiting the present invention.

[0012] Please see Figure 1 The method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle provided by the present invention includes steps S1 to S7: Step S1: Obtain vehicle dynamics data through onboard sensors, and establish a multi-axis unmanned differential steering vehicle dynamics model based on the vehicle dynamics data.

[0013] The acquired vehicle dynamics data includes: vehicle longitudinal velocity. Lateral velocity yaw rate Vehicle heading angle Overall vehicle quality Vehicle wheelbase Vehicle No. Distance from the axle to the center of gravity Moment of inertia of the wheel Drive torque Braking torque Tire radius Vehicle No. Longitudinal stiffness of tires on the axle Vehicle No. Lateral stiffness of the tire on the axle .

[0014] First, the dynamic equations for multi-axle vehicles:

[0015]

[0016] in, for The first derivative, For the axle number index, This represents the total number of axles. for The first derivative, For the first The lateral force experienced at the point of contact between the left wheel and the ground on the axle. For the first The lateral force experienced at the point where the right wheel of the axle contacts the ground. For the first The longitudinal force at the contact point of the left wheel of the axle. For the first The longitudinal force experienced at the point where the right wheel of the axle touches the ground.

[0017] When the tire is within its linear range, the longitudinal force of the tire is directly proportional to the longitudinal slip ratio. The longitudinal force of the tire is expressed as:

[0018] in, For left and right side markings, At that time, corresponding to the left wheel, At that time, corresponding to the right wheel, Indicates the first root axle The longitudinal force at the point where the side wheel touches the ground. The total speed difference between the left and right wheels is given by the driver or the control algorithm. At this time, it indicates that the vehicle is turning.

[0019] When the tire is within its linear range, the lateral force of the tire is proportional to the slip angle, and the lateral force of the tire is expressed as:

[0020] in, Indicates the first root axle The lateral force experienced at the point where the side wheel touches the ground.

[0021] Substituting the longitudinal and lateral forces of the tires into the dynamics equations of a multi-axle vehicle allows us to establish a dynamics model for a multi-axle unmanned differential steering vehicle, expressed as follows:

[0022] .

[0023] When given and The answer can be obtained from the above system of equations. , Lateral acceleration Longitudinal acceleration Lateral acceleration differential increment Longitudinal acceleration differential increment This can describe the longitudinal, lateral, and yaw motion characteristics of a multi-axle unmanned differential steering vehicle.

[0024] Step S2: Number the drive actuators according to the arrangement relationship of the drive actuators on the left and right sides of each axle of the vehicle, generate a set of drive actuator failure combinations according to the number of failures and spatial distribution, and build a failure fault library containing each failure combination.

[0025] Specifically, the numbering method for drive actuators is as follows: the vehicle has There are 1 axle, and each axle has a drive actuator on its left and right sides. The drive actuators are numbered as follows: ,in, For the axle number index, The left and right side labels will be provided, and Each wheel position is stored in the actuator failure database, which is shown in Table 1. Table 1

[0026] In Table 1, N represents the fault group number. Multi-axle unmanned differential vehicles exhibit symmetry, allowing fault groups with identical dynamic responses to be grouped together. Specifically, combinations of failures on the same side that do not include all actuators, not mentioned in other tables, are considered fault groups numbered N to 2N-1 based on the maximum interval between actuator numbers; multi-axle and multi-wheel combinations not mentioned in other tables are designated as fault groups numbered 3N+2 to 4N+1; and failure modes with different steering conditions require separate analysis.

[0027] Step S3: For each failure combination in the failure fault library, perform a no-failure baseline simulation and a corresponding failure simulation under preset straight-line driving conditions and preset steering conditions, respectively, to obtain the dynamic response results of the vehicle in the no-failure state and the failure state.

[0028] Specifically, the preset straight-ahead driving condition is as follows: the vehicle is traveling on the road, and the initial speed of the vehicle is set to... After a preset acceleration period, the vehicle speed remained at [a certain value]. And set in the first A specific wheel of the vehicle experiences complete failure within seconds; The preset steering condition is: the vehicle is traveling on the road, and the initial speed of the vehicle is set to... After a preset acceleration period, the vehicle speed remained at [a certain value]. , No. The total speed difference between the left and right wheels is given in seconds. And set in the first The vehicle experiences a complete wheel failure at a specific time.

[0029] Among them, a complete failure fault includes at least one of the following: the drive torque output is set to zero, or the drive torque output is fixed at a preset fault value, or the drive actuator is in an out-of-control state that cannot follow the control command.

[0030] The results are obtained from the dynamic equations of a multi-axle unmanned differential steering vehicle. , , , , , , This can describe the longitudinal, lateral, and yaw motion characteristics of a multi-axle unmanned differential steering vehicle, and thus obtain the dynamic response of the vehicle's longitudinal, lateral, and yaw motion in both the no-failure and failed states.

[0031] Step S4: Based on the dynamic response results under no-failure and failure states, construct the longitudinal severity index, lateral severity index, and yaw severity index.

[0032] The longitudinal severity index is used to characterize the impact of actuator failure on the longitudinal safety of the vehicle, the lateral severity index is used to characterize the impact of actuator failure on the lateral stability of the vehicle, and the yaw severity index is used to characterize the impact of actuator failure on the yaw stability of the vehicle.

[0033] Longitudinal severity index The expression is:

[0034] in, for The differential increment, For the response time of the braking actuator, Longitudinal acceleration The differential increment.

[0035] The longitudinal severity index assesses the collision risk of a vehicle using two parameters: longitudinal acceleration and longitudinal velocity.

[0036] Lateral severity index The expression is:

[0037] in, for The differential increment, To improve the response time of the implementing agency, Lateral acceleration The differential increment.

[0038] The lateral severity index assesses the degree of deviation of a vehicle from its desired path using two parameters: lateral acceleration and lateral velocity.

[0039] Severity index of yaw The expression is:

[0040] in, for The differential increment.

[0041] The yaw severity index assesses the severity of vehicle instability by analyzing the rate of change of yaw angular velocity.

[0042] Step S5: Based on the longitudinal severity index, lateral severity index, and yaw severity index, each actuator failure combination is mapped to a corresponding sub-severity level. The overall severity level of each actuator failure combination is obtained by fusing multiple indices, thereby realizing the severity classification and ranking of chassis actuator failure combinations.

[0043] Specifically, based on the value ranges of the longitudinal severity index, lateral severity index, and yaw severity index, each sub-severity level is divided into multiple levels, and the overall severity level is determined by the fusion result of the sub-severity levels.

[0044] Define the longitudinal sub-severity level, lateral sub-severity level, and yaw sub-severity level as follows: , , .

[0045] Then, the overall severity score The expression is: .

[0046] After the simulation, the severity ranking table for different failure conditions is calculated based on the obtained dynamic response parameters, as shown in Table 2. The table can be used as an actuator failure evaluation table.

[0047] Table 2

[0048] Based on the overall severity score The overall severity level was determined, as shown in Table 3.

[0049] Table 3

[0050] Step S6: Analyze the reliability of the drive system based on the failure database. Construct a fault tree with the failure of a single drive actuator as the basic event and the number of failures exceeding the tolerance limit as the intermediate event. Calculate the system failure probability under both fault-tolerant and fault-tolerant control structures.

[0051] For multi-axle vehicle drive systems, the failure of each motor is treated as an independent basic event. The calculation of the system failure probability primarily depends on the number of hub motors and the maximum number of motors that can withstand failure after adding fault-tolerant control. A binomial distribution is used to calculate the system failure probability. Specifically, the system failure probability... for:

[0052] in, For the total number of driving actuators, Tolerance limit, This represents the number of actual failed drive actuators. Indicates from Select from the drive actuators The number of combinations of failed drive actuators. To ensure the reliability of the actuator, The failure rate of the actuator.

[0053] Step S7: Make a comprehensive risk judgment based on the overall severity level and the corresponding system failure probability, and output the basis for vehicle operation constraints and / or fault-tolerant control intervention corresponding to different risk levels.

[0054] Specifically, by pre-setting risk judgment rules, the overall severity level and the system failure probability are jointly mapped to the final risk level R. The risk judgment rules are shown in Table 4. This represents the minimum probability of system failure. In this embodiment, the maximum system failure probability is defined as... It is 0.005%. The risk level is 0.5%, and the final risk level R is defined as ranging from R0 (no risk) to R4 (extremely high risk).

[0055] Table 4

[0056] Based on the final risk level R, output at least one of the following vehicle operation constraint strategies: speed limit, acceleration limit, yaw rate limit, steering input range limit, or torque redistribution that triggers fault-tolerant control.

[0057] Specifically, based on the final risk level determined by the judgment, the vehicle control system implements the following tiered, progressively escalating response strategy: R0 corresponds to a risk-free strategy: no operational constraints; no control intervention. The vehicle operates at full performance according to the original instructions.

[0058] R1 corresponds to the warning and performance conservative strategy: Operational constraints: Speed ​​limit, restricting the maximum speed to 80% of the current speed or a preset safe speed. Acceleration limit: Limiting the absolute value of driving / braking acceleration to no more than 2.0 m / s². 2 .

[0059] Control intervention: Sends an early warning message to the control system that "actuator failure, performance limitation".

[0060] R2 corresponds to the stability guarantee and basic fault tolerance strategy: Operational constraints: Enhanced lateral constraints limit the absolute value of lateral acceleration to no more than 0.3g and the yaw rate to no more than 0.5 rad / s. Steering difference is limited by the total speed difference between the left and right wheels. Limit the amplitude to prevent excessive yaw moment.

[0061] Control intervention: Triggers the torque redistribution fault-tolerant control algorithm to redistribute torque within the capabilities of the healthy actuator to compensate as much as possible for the power loss and torque imbalance caused by the failed actuator. A message appears: "Stability control is activated."

[0062] R3 corresponds to high-risk intervention and pathway planning strategies: Operational constraints: Forced deceleration at -0.5 m / s 2 The deceleration smoothly reduces the vehicle speed to a low speed (e.g., 20 km / h). Steering input lock takes over steering control, locking out drastic changes in driver input, and the fault-tolerant controller maintains the vehicle's basic heading.

[0063] Control intervention: Triggers integrated fault-tolerant control and local path replanning. Combining high-precision maps and perception information, it calculates and guides the vehicle to the nearest emergency lane or safe parking area. It requests driver takeover or prepares for remote intervention.

[0064] R4 corresponds to the minimum risk state strategy: Operational constraints and intervention: Emergency safety stop, triggering "minimum risk management", and performing emergency braking with maximum deceleration until the vehicle stops, while ensuring the vehicle's attitude is stable (such as through differential braking).

[0065] Power isolation: Cuts off power output to the failed side or all drive actuators, retaining only braking and basic steering functions. Hazard warning lights are activated via hazard lights and an emergency alarm is sent to the backend containing location, fault type, and risk level.

[0066] Verification case: such as Figures 2 to 8 The image illustrates an exemplary road test scenario. The test vehicle is a three-axle unmanned vehicle with differential steering. The preset straight-line driving condition is specifically set on a road surface, with the vehicle's initial speed set to [value missing]. =5m / s, after a certain period of acceleration, the vehicle speed remains at... =10m / s, and it is set that a specific wheel of the vehicle will completely fail at the 10th second. The preset turning condition is specifically set on road driving, and the initial speed of the vehicle is set to... =5m / s, after a certain period of acceleration, the vehicle speed remains at... =10m / s, the total speed difference between the left and right wheels is given at the 5th second. It is set that a specific wheel of the vehicle will completely fail at the 10th second.

[0067] Table 1 categorizes the number of actuator failures in the three-axle unmanned vehicle, selecting typical operating conditions including single-wheel failure, two-wheel failure, three-wheel failure, and failures on the same side, coaxial, and diagonal. Conditions requiring shutdown due to multiple actuator failures and some symmetrical conditions are excluded. The resulting chassis actuator failure database for the three-axle vehicle is shown in Table 5.

[0068] Table 5

[0069] Under steady-state straight-line driving conditions, considering the symmetry of the wheels, the left front wheel, left middle wheel, and left rear wheel are selected to represent a single-wheel failure straight-line fault. Therefore, fault groups 1, 2, and 3 are selected for the experiment. The vehicle dynamic response under single-wheel drive system failure and no-fault conditions is as follows: Figure 2 As shown, Figure 2 In the table, demo 0, demo 1, demo 2, and demo 3 correspond to the fault modes numbered 0, 1, 2, and 3 of the straight-line fault group shown in Table 5, respectively.

[0070] Depend on Figure 2 It can be seen that when the vehicle's drive system does not fail, the vehicle maintains steady straight-line travel without lateral displacement, and both lateral velocity and yaw rate remain constant at zero. Due to road surface excitation and aerodynamic effects, vehicle parameters exhibit slight fluctuations at the initial start-up. When the vehicle's drive system fails at the tenth second, the vehicle stability parameters all change significantly. Within 50 seconds of the drive system failure, the lateral displacement of the vehicle under the above conditions deviates by more than 80 meters. When the front axle motor fails, the lateral force generates a clockwise yaw moment around the center of gravity, offsetting part of the yaw moment generated by the longitudinal force. However, when the rear axle motor fails, the effect is exactly the opposite. Therefore, compared to the failure of the front and middle axle motors, the failure of the rear axle motor causes more pronounced yaw instability and lane departure.

[0071] Six-wheel differential vehicles driven by hub motors are typical overdrive vehicles with redundancy. Since unmanned vehicles control speed via algorithms, when a motor fails, the output torque is increased to meet vehicle power demands, exceeding the motor's maximum output torque. However, a slight difference remains compared to the vehicle speed before the failure. Because the torque to each wheel is evenly distributed across the total torque in straight-line driving conditions, the power loss is the same in conditions 1, 2, and 3. When extended to multi-axle vehicles, the impact on yaw stability is greater when a single-wheel failure occurs on a tire closer to the rear axle. This is because the yaw moment generated by the lateral force lost due to a rear axle motor failure is the same as the yaw moment generated by the lost longitudinal force. Furthermore, the more axles there are and the farther the rear axle is from the center of gravity, the greater the resulting yaw moment. However, for overall vehicle power, the power loss is the same for each wheel.

[0072] Under steady-state straight-line driving conditions, the failure of the co-axle wheel drive system (i.e., front axle wheel failure, middle axle wheel failure, and rear axle wheel failure) only resulted in a loss of dynamic performance in the vehicle's dynamic response during the experiment. Therefore, for combined fault groups 6, 8, and 9, only the vehicle's longitudinal velocity and longitudinal acceleration were analyzed. The experimental results are as follows: Figure 3 As shown, Figure 3In the table, demo 6, demo 8, and demo 9 correspond to the fault modes numbered 6, 8, and 9 in the straight-line fault group shown in Table 5.

[0073] Through analysis Figure 3 It is known that for vehicles with differential steering, the vehicle steers by the speed difference between the two sides, and in straight-line driving conditions, the torque of each tire is evenly distributed. Therefore, when two tires on one axle fail simultaneously, it will not affect the vehicle's yaw stability, and the loss of vehicle power will be the same. Similarly, for multi-axle vehicles, the failure of the coaxial motor will not significantly affect the vehicle's yaw stability, and the loss of power will also be the same.

[0074] During steady-state straight-line driving, the number of wheel motor failures significantly impacts the vehicle's dynamic response. When all three motors on one side of the vehicle fail, the vehicle can no longer operate normally. Therefore, this section selects combined fault groups 1, 4, and 10 for comparative experiments. The experimental results are as follows: Figure 4 As shown, Figure 4 In the table, demo 1, demo 4, and demo 10 correspond to the fault modes numbered 1, 4, and 10 of the straight-line fault group shown in Table 5, respectively.

[0075] from Figure 4 As can be seen, when the drive motors begin to fail at the 10-second mark, the dynamic responses under all three operating conditions change significantly, with the order being: three-wheel failure > two-wheel failure > single-wheel failure. Analysis reveals that compared to the failure of a single drive motor, the failure of multiple drive motors results in worse yaw stability and greater power loss. When multiple drive motors fail on the same side of the vehicle, the yaw rate is significantly disturbed, leading to greater lateral displacement and more severe consequences during driving. For multi-axle vehicles, the more motors that fail, the greater the power loss. When motor failures occur primarily on one side of the vehicle, the more wheels fail, resulting not only in greater power loss but also worse yaw stability. Especially in multi-axle vehicles where all motors on one side fail, the vehicle becomes unusable, leading to extremely serious consequences.

[0076] When the vehicle is traveling in a steady straight-line condition, the dynamic response of the vehicle deviates significantly from the normal driving condition when the motors of the co-axle wheels, the diagonal wheels, and the same-side wheels fail. In all cases, two wheels fail. Therefore, fault groups 5, 6, and 7 were selected for comparative experiments. The experimental results are as follows: Figure 5 As shown, Figure 5 In the table, demo 5, demo 6, and demo 7 correspond to the fault modes numbered 5, 6, and 7 of the straight-line fault group shown in Table 5, respectively.

[0077] Depend on Figure 5 It can be seen that when the motor fails at the 10-second mark, the instability caused by the failure of the motor on the same side is more severe than that caused by the failure of the diagonal motor or the coaxial motor. This is because the failure of the motor on the same side creates a torque difference between the two sides of the vehicle, resulting in a yaw moment and instability. In contrast, the failure of the motor on the same side and the diagonal motor do not create a torque difference between the two sides of the vehicle. Regarding vehicle dynamics, since the test conditions all involved two-wheel failure, the difference in power loss is not significant. However, the longitudinal acceleration of the two wheels on the same side after failure will fluctuate around zero, reducing vehicle comfort. Based on the above analysis and experimental data, when the number of failed motors is the same, the stability of multi-axle vehicles with motors failing on opposite sides is better than those with motors failing on the same side. This is because the failure of the motor on the same side creates a larger torque difference between the two sides of the vehicle, resulting in a larger yaw moment and more deviation when the vehicle is traveling in a straight line.

[0078] When analyzing three-wheel failures, due to the excessive number of vehicle drive motor failures and excluding the extreme case of complete motor failure on one side of the vehicle, failure conditions 11 and 12 were selected for comparative experiments. The experimental results are as follows: Figure 6 As shown, Figure 6 In the table, demo 11 and demo 12 correspond to the fault modes numbered 11 and 12 in the straight-line fault group shown in Table 5, respectively.

[0079] Depend on Figure 6 It can be seen that when all three motors fail, the rear axle motor failure has more severe consequences than the front axle motor failure, resulting in a greater yaw rate and a more pronounced lateral displacement. Due to the difference in load on the front and rear axles and the different motor loads, the yaw moment imbalance is more severe when the rear axle motor fails, leading to more obvious yaw instability and deviation. In terms of dynamics, since the torque distribution method of the wheels is to distribute the total torque evenly to each motor, and the test conditions are all three-wheel failure, the total longitudinal force loss remains unchanged, so the difference in dynamics loss under the test conditions is not significant; however, the longitudinal acceleration when the rear axle motor fails will oscillate more around zero after failure, resulting in a greater reduction in vehicle comfort.

[0080] Based on the above, the severity of failures in the drive system under straight-line operating conditions is ranked, resulting in longitudinal severity ranking tables, lateral severity ranking tables, yaw severity ranking tables, and overall severity level tables, as shown in Tables 6 to 9, respectively: Table 6

[0081] Table 7

[0082] Table 8

[0083] Table 9

[0084] By analyzing and comparing the experimental data, under the set steady-state straight driving conditions, if a drive system failure occurs, the failure of the three-wheel drive system on the same side is extremely dangerous, followed by the failure of the two-wheel drive system on the same side; diagonal failure and coaxial failure have less impact on the driver; for three-wheel failure, compared with conditions 11 and 12, the impact of front axle failure is less than that of rear axle failure; for single-wheel drive system failure, rear wheel failure is more stable than front wheel failure, but the loss of power is the same.

[0085] Vehicle dynamics analysis was conducted under turning conditions, including single-wheel drive system failure and multi-wheel system failure on the same side.

[0086] Due to the symmetry of the differential steering vehicle structure, experiments were conducted on the turning conditions with failure of the right-hand drive system (first, second, and third wheels) and the dynamic response of the affected vehicles were compared with that of vehicles without faults. Figure 7 As shown. Figure 7 In the table, demo 0, demo 1, demo 2, and demo 3 correspond to the fault modes numbered 0, 1, 2, and 3 of the turning fault group shown in Table 5, respectively.

[0087] Depend on Figure 7 It can be seen that when the vehicle is functioning correctly, it initially accelerates to 10 m / s, then begins to steer after 5 seconds, achieving steady-state steering after a certain period, with a lateral speed maintained at -0.8 m / s. When a single-wheel drive system fails, the vehicle's dynamic response changes; however, because failure of the front and rear axle drive systems generates a yaw moment at the center of gravity, the vehicle experiences fishtailing, leading to understeer or oversteer. Therefore, the impact of front and rear axle failure is greater than that of mid-axle drive system failure. Since each tire receives the same longitudinal force, the vehicle's power loss is the same. Similarly, for multi-axle vehicles, the farther the axle is from the wheels, the greater the yaw moment generated at the center of gravity when the motor on that axle fails, and the greater the impact on the vehicle's lateral stability.

[0088] Experiments were conducted to compare the dynamic responses of a fault-free vehicle with a multi-wheel drive system failure on the same side under steady-state steering conditions. Figure 8 As shown. Figure 8 In the table, demo 1, demo 4, and demo 10 correspond to the fault modes numbered 1, 4, and 10 of the turning fault group shown in Table 5, respectively.

[0089] The data analysis above shows that, compared to single-wheel drive system failure, failure of a multi-wheel drive system on the same side has a greater impact on the vehicle, resulting in more severe skidding. Furthermore, the order of severity is: three-wheel failure > two-wheel failure > single-wheel failure. For a three-wheel drive system failure on the same side, this will have serious consequences for the driver, requiring immediate cessation of driving. For multi-axle vehicles, the more wheels that fail, the greater the impact on lateral stability. Especially when all motors on the same side fail, the vehicle will be unable to move normally, posing a significant danger and requiring immediate stopping and repair.

[0090] The severity of the dynamic response to wheel drive system failures under steering conditions is ranked according to Table 1, as shown in Tables 10 to 13: Table 10

[0091] Table 11

[0092] Table 12

[0093] Table 13

[0094] Analysis of the above data reveals that when a single-wheel drive system fails, the severity of the failure is higher for the front and rear axle tires than for the center axle tires. For six-wheel differential steering vehicles, the yaw moment generated by a failed front or rear axle tire during cornering, leading to a more pronounced fishtailing phenomenon, makes the impact of a front or rear axle tire failure on the vehicle's overall severity more significant. Furthermore, when a multi-wheel drive system on the same side fails, the danger level of each fault group is very high, especially in the case of a failed three-wheel drive system on the same side. This situation is extremely dangerous and requires immediate stopping and repair.

[0095] To conduct quantitative analysis of the system, a three-axle differential steering vehicle driven by a hub motor was selected for analysis. Through analysis, the reliability of the system under different failure states can be classified according to different failure probabilities.

[0096] With the addition of fault-tolerant control, the power loss of the vehicle can be compensated by utilizing the redundancy of the overdrive vehicle and redistributing the wheel torque through the fault-tolerant control strategy. However, if all three motors on one side of the vehicle fail, the vehicle system is also considered to have failed. Therefore, the maximum number of motors that can be tolerated for failure after the addition of fault-tolerant control can generally be 3, but the two cases of all motors on the same side of the vehicle failing must be excluded.

[0097] Therefore, based on the above analysis, the fault tree diagram with added fault-tolerant control is drawn as follows: Figure 9 As shown. Figure 9 In the text, X1, X2, X3, X4, X5, and X6 represent the numbers respectively. , , , , , The actuators fail individually; in systems with fault-tolerant control, failures are caused by three intermediate events: failure of the left three wheels, failure of the right three wheels, and failure of at least four wheels, connected by an OR gate. Failure of the three wheels on the same side is caused by three failed motors on the same side, connected by an AND gate. Failure of at least four wheels is caused by three intermediate events: failure of four wheels, failure of five wheels, and failure of six wheels, connected by an OR gate. These result in 15, 6, and 1 types of motor failure scenarios, respectively. After the above qualitative analysis, the system will be quantitatively analyzed next.

[0098] Setting the failure rate of the drive actuator Then reliability .

[0099] The failure probability of a fault-tolerant control system is :

[0100] Assuming a fault tree with fault-tolerant control Figure 3 The intermediate events are left-side three-wheel failure, right-side three-wheel failure, and at least four-wheel failure, respectively. , , Then the failure probability of a fault-tolerant control system is :

[0101] Among them, the event probability .

[0102] event probability .

[0103] event probability .

[0104] Calculate the intersection probability to exclude duplicate counts: Probability of failure of all six motors .

[0105] Motors 1, 2, and 3 have failed, and at least four motors in total have failed, meaning that at least one of the right-side motors (4, 5, and 6) has failed. .

[0106] Motors 4, 5, and 6 have failed, and at least four motors in total have failed, meaning that at least one of the left-side motors (1, 2, and 3) has failed. .

[0107] Motors 1, 2, and 3 fail (A), motors 4, 5, and 6 fail (B), and at least four motors fail (C). Since A and B imply the failure of all six motors, C is inevitable. Therefore, .

[0108] Therefore, the failure probability of a fault-tolerant control system is also known as the system failure probability. .

[0109] The ratio of the failure probability to the system failure probability of a fault-tolerant control system is: .

[0110] It can be seen that the failure probability of a three-axis unmanned differential vehicle with fault-tolerant control is about 0.0002%, while that without fault tolerance is about 5.8%. It can be concluded that adding fault-tolerant control can effectively reduce the overall failure rate of the three-axis unmanned vehicle system, and it is necessary to integrate fault-tolerant control into vehicle control.

[0111] Based on the overall severity level and system failure probability of the two failure scenarios (straight-going and turning), and combined with the final risk levels of R0-R4, the following conclusions can be drawn: Under straight-line driving conditions, the final risk level for failures numbered 1, 6, 7, 8, and 9 is R0, with a no-risk strategy: no operational constraints; no control intervention. The vehicle operates at full performance according to the original instructions.

[0112] Under straight-traffic conditions, the final risk level for failures numbered 2, 3, and 11 is R1. The selected warning and performance conservative strategy is: Operational constraints: Speed ​​limit, restricting the maximum vehicle speed to 80% of the current speed or a preset safe speed; Acceleration limit: limiting the absolute value of drive / braking acceleration to no more than 2.0 m / s². 2 Control intervention: Sends an early warning message to the control system that "actuator failure, performance limitation".

[0113] Under straight-line driving conditions, the final risk level for failure scenarios numbered 4, 5, and 12 is R2. The selected stability assurance and basic fault-tolerance strategy is as follows: Operational constraints: Enhanced lateral constraints: Limiting the absolute value of lateral acceleration to no more than 0.3g and the yaw rate to no more than 0.5 rad / s. Steering difference limitation: Limiting the total speed difference between the left and right wheels. Limit the amplitude to prevent excessive yaw moment.

[0114] Control intervention: Triggers the torque redistribution fault-tolerant control algorithm to redistribute torque within the capabilities of the healthy actuator to compensate as much as possible for the power loss and torque imbalance caused by the failed actuator. A message appears: "Stability control is activated."

[0115] Under straight-line driving conditions, the final risk level of failure case number 10 is R4. The minimum risk state strategy is selected: Operational constraints and interventions: Emergency safe stop: Trigger "Minimum Risk Management," and, while ensuring vehicle stability (e.g., through differential braking), execute emergency braking with maximum deceleration until the vehicle stops; Power isolation: Cut off the power output of the failed side or all drive actuators, retaining only braking and basic steering functions; Hazard warning lights and remote alarm: Activate hazard warning lights and send an emergency alarm to the backend containing location, fault type, and risk level.

[0116] The final risk level for failure scenario number 2 under turning conditions is R0, with a no-risk strategy: no operational constraints; no control intervention. The vehicle operates at full performance according to the original instructions.

[0117] Under turning conditions, the final risk level for failures numbered 1 and 3 is R1. The selected warning and performance conservative strategy is: Operational constraints: Speed ​​limit, restricting the maximum vehicle speed to 80% of the current speed or a preset safe speed; Acceleration limit: limiting the absolute value of drive / braking acceleration to no more than 2.0 m / s². 2 Control intervention: Sends an early warning message to the control system that "actuator failure, performance limitation".

[0118] Failure scenario number 4 under cornering conditions has a final risk level of R2. The selected stability assurance and basic fault tolerance strategy is as follows: Operational constraints: Enhanced lateral constraints: Limiting the absolute value of lateral acceleration to no more than 0.3g and the yaw rate to no more than 0.5 rad / s. Steering difference limitation: Limiting the total speed difference between the left and right wheels. Limit the amplitude to prevent excessive yaw moment.

[0119] Control intervention: Triggers the torque redistribution fault-tolerant control algorithm to redistribute torque within the capabilities of the healthy actuator to compensate as much as possible for the power loss and torque imbalance caused by the failed actuator. A message appears: "Stability control is activated."

[0120] Under turning conditions, the final risk level of failure case number 10 is R4. The minimum risk strategy is selected: Operational constraints and interventions: Emergency safe stop: Trigger "Minimum Risk Management," and, while ensuring vehicle stability (e.g., through differential braking), execute emergency braking with maximum deceleration until the vehicle stops; Power isolation: Cut off power output to the failed side or all drive actuators, retaining only braking and basic steering functions; Hazard warning lights and remote alarm: Activate hazard warning lights and send an emergency alarm to the backend containing location, fault type, and risk level.

[0121] In summary, the actuator failure assessment and reliability analysis method for multi-axle unmanned differential vehicles according to the above embodiments has the following beneficial effects: (1) By numbering the drive actuators and generating a set of failure combinations according to the number of failures and spatial distribution, and constructing a failure database, the present invention can systematically organize and uniformly manage the failure combinations of drive actuators of multi-axle unmanned differential vehicles, effectively solving the problem that the existing technology is difficult to cover failure combinations with significant differences in spatial distribution of multi-axle vehicles.

[0122] (2) The present invention conducts a failure-free benchmark simulation and a corresponding failure simulation under a preset straight-going condition and a preset turning condition, respectively, to obtain the dynamic response under the failure-free state and the failure state, so that the failure effect has a consistent reference.

[0123] (3) Based on dynamic response, the present invention constructs longitudinal severity index, lateral severity index and yaw severity index, maps failure combinations to sub-severity levels, and obtains the overall severity level through multi-index fusion, effectively realizing the severity classification and ranking of chassis actuator failure combinations.

[0124] (4) This invention performs drive system reliability analysis based on a failure database. It constructs a fault tree with the failure of a single drive actuator as the basic event and the number of failures exceeding the tolerance limit as the intermediate event. It calculates the system failure probability under fault-tolerant control structure and fault-tolerant control structure respectively, and makes a comprehensive risk judgment in combination with the overall severity level. It can output vehicle operation constraints and fault-tolerant control intervention basis corresponding to different risk levels, and solves the problem that the existing technology is difficult to form a comprehensive risk judgment and operation constraint output for fault-tolerant control intervention.

[0125] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle, characterized in that, include: Step S1: Obtain vehicle dynamics data through onboard sensors, and establish a multi-axle unmanned differential steering vehicle dynamics model based on the vehicle dynamics data; Step S2: Number the drive actuators according to the arrangement relationship of the drive actuators on the left and right sides of each axle of the vehicle, generate a set of drive actuator failure combinations according to the number of failures and spatial distribution, and build a failure fault library containing each failure combination. Step S3: For each failure combination in the failure database, perform a no-failure baseline simulation and a corresponding failure simulation under preset straight-line driving conditions and preset steering conditions, respectively, to obtain the dynamic response results of the vehicle in the no-failure state and the failure state. Step S4: Based on the dynamic response results under no-failure and failed states, construct the longitudinal severity index, lateral severity index, and yaw severity index. Step S5: Based on the longitudinal severity index, lateral severity index, and yaw severity index, each actuator failure combination is mapped to a corresponding sub-severity level, and the overall severity level of each actuator failure combination is obtained through multi-index fusion, thereby realizing the severity classification and ranking of chassis actuator failure combinations. Step S6: Analyze the reliability of the drive system based on the failure database. Construct a fault tree with the failure of a single drive actuator as the basic event and the number of failures exceeding the tolerance limit as the intermediate event. Calculate the system failure probability under the fault-tolerant control structure and the fault-tolerant control structure respectively. Step S7: Make a comprehensive risk judgment based on the overall severity level and the corresponding system failure probability, and output the basis for vehicle operation constraints and / or fault-tolerant control intervention corresponding to different risk levels.

2. The method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle according to claim 1, characterized in that, In step S1, the acquired vehicle dynamics data includes: vehicle longitudinal velocity. Lateral velocity yaw rate Vehicle heading angle Overall vehicle quality Vehicle wheelbase Vehicle No. Distance from the axle to the center of gravity Moment of inertia of the wheel Drive torque Braking torque Tire radius Vehicle No. Longitudinal stiffness of tires on the axle Vehicle No. Lateral stiffness of the tire on the axle ; The expression for the established dynamics model of the multi-axis unmanned differential steering vehicle is as follows: in, for The first derivative, For the axle number index, This represents the total number of axles. for The first derivative, This represents the total speed difference between the left and right wheels.

3. The method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle according to claim 2, characterized in that, In step S3, the preset straight-ahead driving condition is as follows: the vehicle is traveling on the road, and the initial speed of the vehicle is set to... After a preset acceleration period, the vehicle speed remained at [a certain value]. And set in the first A specific wheel of the vehicle experiences complete failure within seconds; The preset steering condition is: the vehicle is traveling on the road, and the initial speed of the vehicle is set to... After a preset acceleration period, the vehicle speed remained at [a certain value]. , No. The total speed difference between the left and right wheels is given in seconds. And set in the first A specific wheel of the vehicle experiences complete failure within seconds; Among them, a complete failure fault includes at least one of the following: the drive torque output is set to zero, or the drive torque output is fixed at a preset fault value, or the drive actuator is in an uncontrolled state that cannot follow the control command.

4. The method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle according to claim 3, characterized in that, In step S4, longitudinal severity indicators The expression is: in, for The differential increment, For the response time of the braking actuator, Longitudinal acceleration The differential increment; Lateral severity index The expression is: in, for The differential increment, To improve the response time of the implementing agency, Lateral acceleration The differential increment; Severity index of yaw The expression is: in, for The differential increment.

5. The method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle according to claim 4, characterized in that, In step S5, the overall severity score is first calculated based on the longitudinal sub-severity level, the lateral sub-severity level, and the yaw sub-severity level. The expression is: in, For vertical sub-severity levels, For lateral severity levels, The severity level of the yaw; Then based on the overall severity score Determine the overall severity level.

6. The method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle according to claim 5, characterized in that, In step S6, the system failure probability for: in, For the total number of driving actuators, Tolerance limit, This represents the number of actual failed drive actuators. Indicates from Select from the drive actuators The number of combinations of failed drive actuators. To ensure the reliability of the actuator, The failure rate of the actuator.

7. The method for actuator failure assessment and reliability analysis of a multi-axle unmanned differential vehicle according to claim 1, characterized in that, Step S7 specifically includes: By pre-setting risk judgment rules, the overall severity level and the system failure probability are jointly mapped to the final risk level, and at least one of the following vehicle operation constraint strategies is output based on the final risk level: speed limit, acceleration limit, yaw rate limit, steering input range limit, or torque redistribution that triggers fault-tolerant control.

Images