Automatic driving path planning and dynamic obstacle avoidance method for electric vehicles

Abstract

The application discloses an automatic driving path planning and dynamic obstacle avoidance method for electric vehicles, and particularly relates to the technical field of vehicle control systems, and is used for solving the problem that the existing automatic driving system does not consider real-time performance attenuation of a driving system when planning a path, thereby leading to control instability; the method is characterized in that: a historical load spectrum of the driving system and a real-time operation condition are acquired in real time, a performance attenuation state is judged based on frequency domain feature coincidence, a heat management strategy direction is generated in combination with thermal inertia characteristics and obstacle avoidance urgency, an action feasibility space that meets both internal dynamics constraints and external space constraints is constructed, a system performance constraint evaluation result is obtained by analyzing a dynamic envelope, a real-time maximum output capability boundary of the driving system is predicted, a dynamic obstacle avoidance path is replanned based on the boundary, and finally the vehicle is controlled to perform an obstacle avoidance operation, so that dynamic matching of a physical execution capability of the vehicle and a path planning requirement is realized, and the executability of the obstacle avoidance path within the actual capability range of the vehicle is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle control systems, and more particularly, to an automatic driving path planning and dynamic obstacle avoidance method for electric vehicles. BACKGROUND

[0002] In the process of path planning and dynamic obstacle avoidance of existing electric vehicle automatic driving systems, trajectory generation and control decisions are usually made based on the performance model of the vehicle in an ideal state, dynamic obstacles are identified through an environment perception module, and an obstacle avoidance path is planned according to preset vehicle dynamics parameters, and finally tracking operations are completed through a drive-by-wire execution mechanism. This kind of method generally assumes that the drive system, steering system and braking system can continuously provide the required power output and response performance for planning.

[0003] However, in actual operation, especially in continuous or extreme obstacle avoidance conditions, the electric drive system of an electric vehicle is prone to heat accumulation due to high power output, which may cause system output performance degradation, thus making the dynamic obstacle avoidance path planned based on the ideal performance model unable to be effectively tracked, and there is a risk of vehicle control instability and obstacle avoidance failure. The existing method lacks a dynamic perception and fusion mechanism for the real-time physical execution capability boundary of the vehicle, and it is difficult to guarantee the feasibility and safety of the path planning result in all working conditions. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides an automatic driving path planning and dynamic obstacle avoidance method for electric vehicles to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0006] The automatic driving path planning and dynamic obstacle avoidance method for electric vehicles comprises:

[0007] S1, real-time acquisition of historical load spectrum, real-time operating condition and vehicle operating environment data of the electric vehicle drive system;

[0008] S2, judging whether the drive system is in or about to enter a performance degradation state according to the frequency domain feature coincidence degree of the historical load spectrum and the real-time operating condition;

[0009] S3, when in or about to enter the performance degradation state, performing time domain trade-off evaluation based on the thermal inertia characteristics of the drive system and the urgency of the obstacle avoidance action to generate a thermal management strategy guide;

[0010] S4, constructing an action feasibility space that meets both internal dynamics constraints and external space constraints according to the thermal management strategy guide and the vehicle operating environment data, and obtaining a system performance constraint evaluation result by analyzing the dynamic envelope thereof;

[0011] S5, predicting a real-time maximum output capability boundary of the driving system according to the evaluation result of the system performance constraint;

[0012] S6, replanning a dynamic obstacle avoidance path according to the real-time maximum output capability boundary and the vehicle operating environment data, and controlling the electric vehicle to perform an obstacle avoidance operation.

[0013] Further, the historical load spectrum of the driving system of the electric vehicle, the real-time operating condition and the vehicle operating environment data are acquired in real time, including:

[0014] acquiring the pre-stored load spectrum data of the driving system in the historical operating period;

[0015] collecting the torque command, the actual output torque and the speed of the driving motor at the current time as the real-time operating condition;

[0016] receiving the position, speed and road geometry information of the surrounding obstacles output by the environment perception sensor as the vehicle operating environment data.

[0017] Further, according to the frequency domain feature coincidence degree of the historical load spectrum and the real-time operating condition, it is judged whether the driving system is in or about to enter the performance attenuation state, including:

[0018] the historical load spectrum and the real-time operating condition are respectively subjected to frequency domain transformation to obtain corresponding historical frequency domain features and real-time frequency domain features;

[0019] the distribution similarity of the historical frequency domain features and the real-time frequency domain features in the key frequency band is calculated as the frequency domain feature coincidence degree: the power spectral density distribution of the historical and real-time operating condition signals in the main energy frequency band range is extracted; the power spectral density distribution correlation coefficient of the historical and real-time operating condition signals in the same key frequency band is calculated; and the power spectral density distribution correlation coefficient is taken as the quantitative index of the frequency domain feature coincidence degree;

[0020] If the frequency domain feature coincidence degree is lower than a preset coincidence threshold, it is determined that the driving system is in or about to enter the performance attenuation state.

[0021] Further, when in or about to enter the performance attenuation state, a time domain trade-off evaluation is performed based on the thermal inertia characteristic of the driving system and the urgency of the obstacle avoidance action to generate a thermal management strategy guide, including:

[0022] the thermal relaxation time of the driving system is estimated based on the thermal capacity and thermal resistance parameters to represent the thermal inertia characteristic;

[0023] the obstacle avoidance action time window is calculated according to the relative motion relationship of the dynamic obstacles in the vehicle operating environment data to represent the urgency of the obstacle avoidance action;

[0024] the ratio of the thermal relaxation time to the obstacle avoidance action time window is taken as the time domain trade-off factor;

[0025] According to the continuous numerical interval where the time domain trade-off factor is located, a heat management strategy guide that smoothly transitions between heat protection and obstacle avoidance performance is dynamically generated.

[0026] Further, the smaller the time domain trade-off factor, the more the heat management strategy guide tends to prioritize heat protection, and the larger the time domain trade-off factor, the more the heat management strategy guide tends to prioritize obstacle avoidance performance.

[0027] Further, according to the heat management strategy guide and vehicle operating environment data, an action feasibility space that meets both internal dynamics constraints and external space constraints is constructed, and a system performance constraint evaluation result is obtained by analyzing its dynamic envelope, including:

[0028] Mapping the heat management strategy guide to the constraint range of the vehicle's longitudinal and lateral acceleration to form internal dynamics constraints;

[0029] Generating the geometric boundary of the vehicle's drivable area according to the obstacle position, road boundary and traffic rules in the vehicle operating environment data to form external space constraints;

[0030] Establishing a multi-constraint optimization problem with internal dynamics constraints as inequality constraints and external space constraints as equality constraints;

[0031] By real-time iterative solving of the feasible solution set of the multi-constraint optimization problem, an action feasibility space that updates in real time with vehicle motion is constructed;

[0032] Analyzing the coupling relationship between the envelope boundary curvature change and the envelope volume shrinkage rate in the predicted time domain;

[0033] Based on the collaborative evolution mode of the envelope boundary curvature change and the envelope volume shrinkage rate, the system performance constraint evaluation result is determined.

[0034] Further, by real-time iterative solving of the feasible solution set of the multi-constraint optimization problem, an action feasibility space that updates in real time with vehicle motion is constructed, including: initializing the current vehicle state initial value of the multi-constraint optimization problem at each control period; using a numerical optimization algorithm to solve the vehicle state set that meets the internal dynamics inequality constraints and the external space equality constraints; and taking the feasible solution set obtained as the action feasibility space at the current time.

[0035] Further, based on the collaborative evolution mode of the envelope boundary curvature change and the envelope volume shrinkage rate, the system performance constraint evaluation result is determined, including: monitoring the time series data of the envelope boundary curvature change rate and the envelope volume shrinkage rate; establishing a time domain correlation model between the envelope boundary curvature change rate and the envelope volume shrinkage rate; when the collaborative deterioration mode of intensified curvature change and accelerated volume shrinkage is detected, the evaluation result of intensified system performance constraints is generated.

[0036] Further, predicting the real-time maximum output capability boundary of the driving system according to the system performance constraint evaluation result, comprising:

[0037] Extracting the envelope boundary curvature change and envelope volume shrinkage rate in the system performance constraint evaluation result;

[0038] Mapping the envelope boundary curvature change to the maximum torque limit of the sustainable output of the driving system;

[0039] Mapping the envelope volume shrinkage rate to the maximum speed change rate limit allowed by the driving system;

[0040] Combining the maximum torque limit and the maximum speed change rate limit, generating the real-time maximum output capability boundary of the driving system in the predicted time domain.

[0041] Further, re-planning the dynamic obstacle avoidance path according to the real-time maximum output capability boundary and the vehicle operating environment data, and controlling the electric vehicle to perform obstacle avoidance operation, comprising:

[0042] Generating a speed profile that meets the instantaneous power limit of the driving system based on the real-time maximum output capability boundary;

[0043] Combining the obstacle position and road geometry information in the vehicle operating environment data, planning an obstacle avoidance trajectory that meets the kinematics of the vehicle under the constraint of the speed profile;

[0044] Converting the obstacle avoidance trajectory into driving motor torque instructions and steering system angular displacement instructions;

[0045] Sending corresponding instructions to the driving motor controller and the steering system controller through the vehicle bus to execute the obstacle avoidance operation.

[0046] Compared with the prior art, the present application has the following beneficial effects:

[0047] 1. Through real-time monitoring and performance prediction of the driving system operating state, the actual execution capability of the vehicle is accurately perceived, the performance degradation trend of the driving system is identified in advance by analyzing the frequency domain feature coincidence degree of the historical load spectrum and the real-time operating condition, and the time domain weighting evaluation is combined with the thermal inertia characteristic and the obstacle avoidance urgency to generate adaptive thermal management strategy, based on the dynamic perception mechanism of physical execution capability boundary, the control instability problem caused by system performance degradation is effectively avoided.

[0048] 2. By constructing the action feasibility space that meets the internal dynamics constraints and external space constraints, the dynamic matching of vehicle execution capability and path planning requirements is realized, the real-time maximum output capability boundary predicted based on the system performance constraint evaluation result provides accurate physical constraint conditions for path planning, so as to ensure that the generated obstacle avoidance path is executable within the current actual capability range of the vehicle, and the safety and reliability of the automatic driving system under complex working conditions are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 The flowchart of the automatic driving path planning and dynamic obstacle avoidance method for electric vehicles of the present application. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0051] Embodiment: Figure 1 The automatic driving path planning and dynamic obstacle avoidance method for electric vehicles of the present application is given, which comprises:

[0052] S1, real-time acquisition of historical load spectrum, real-time operating condition and vehicle operating environment data of the electric vehicle driving system;

[0053] S2, judging whether the driving system is in or about to enter a performance attenuation state according to the frequency domain feature coincidence degree of the historical load spectrum and the real-time operating condition;

[0054] S3, when in or about to enter the performance attenuation state, performing time domain trade-off evaluation based on the thermal inertia characteristics of the driving system and the urgency of obstacle avoidance action, and generating a thermal management strategy guide;

[0055] S4, constructing an action feasibility space that meets the internal dynamics constraints and external space constraints according to the thermal management strategy guide and the vehicle operating environment data, and obtaining system performance constraint evaluation results by analyzing the dynamic envelope thereof;

[0056] S5, predicting the real-time maximum output capability boundary of the driving system according to the system performance constraint evaluation results;

[0057] S6, re-planning a dynamic obstacle avoidance path according to the real-time maximum output capability boundary and the vehicle operating environment data, and controlling the electric vehicle to perform obstacle avoidance operation.

[0058] In the process of acquiring the historical load spectrum, real-time operating condition and vehicle operating environment data of the electric vehicle driving system in real time, the first step is to acquire the pre-stored load spectrum data of the driving system in the historical operating period. The historical operating period here usually refers to the data period recorded by the vehicle in the past continuous operating time, for example, the past 7 days or 30 days can be selected as the typical historical data period. The load spectrum data specifically refers to the statistical distribution of a series of load states experienced by the driving system in the historical operating period, which records the continuous working time and occurrence frequency of the driving motor under different torque and speed combinations. These data are usually pre-stored in the non-volatile memory of the vehicle in the form of a two-dimensional matrix or a data list, such as the EEPROM or Flash memory of the electronic control unit.

[0059] The stored load spectrum data includes but is not limited to the output torque value of the driving motor, the corresponding speed value and the cumulative operating time at each operating point, wherein the unit of torque value is usually Newton-meter (N·m) and the unit of speed value is usually revolutions per minute (r / min). In the actual acquisition process, these historical data are read by accessing the specified memory address or calling the pre-defined data interface function, providing a data basis for subsequent frequency domain analysis. The data verification step is included in the reading process, such as using the cyclic redundancy check (CRC) method to verify the integrity of the data to ensure that the acquired data is accurate and reliable.

[0060] Next, the torque command, actual output torque and speed of the driving motor at the current time are collected as the real-time operating condition. The torque command refers to the target torque value sent by the vehicle controller to the driving motor, which is usually transmitted in the form of a digital signal through the controller area network (CAN) bus; the actual output torque refers to the actual torque value generated by the driving motor, which is obtained through the torque estimation algorithm built-in the motor controller or sensor measurement, and the torque estimation algorithm is usually based on the motor phase current and rotor position information for calculation; the speed refers to the actual rotational speed of the driving motor rotor, which is usually measured by an encoder or a rotary transformer. The collection process is achieved by directly reading the corresponding message data through the vehicle network bus, for example, the torque command value is obtained by parsing from the CAN bus, and the real-time values of the actual output torque and speed are read from the motor controller. These real-time data are collected at a fixed sampling period, typically 10 milliseconds to 100 milliseconds, to ensure that the key features of the driving system dynamic response can be captured. The collected data are pre-processed, including unit unification processing and data validity verification, such as converting the torque value to Newton-meter (N·m) in the International System of Units, converting the speed value to revolutions per minute (r / min), and eliminating abnormal data points that are obviously beyond the physical possible range.

[0061] Finally, the surrounding obstacle position, speed and road geometry information received by the environment perception sensor output are taken as the vehicle operating environment data. The environment perception sensor includes but is not limited to millimeter wave radar, laser radar, visual camera and positioning system. The surrounding obstacle position refers to the coordinate position of the surrounding vehicle, pedestrian and other obstacles relative to the vehicle detected by the sensor, usually expressed in rectangular coordinate system or polar coordinate system, and the coordinate value is usually in meters (m); the speed refers to the relative motion speed of the obstacle, usually in meters per second (m / s); the road geometry information includes lane line curvature, road width, slope and other parameters, wherein the curvature is usually in units of per meter (1 / m), the road width is usually in units of meters (m), and the slope is usually in units of percentage (%). These data are obtained through the output interface of the sensor controller, such as receiving the comprehensive environment data processed by the perception fusion algorithm through Ethernet or CAN bus. The data receiving adopts a periodic interrupt service program or a data callback mechanism to ensure the real-time and accuracy of the environment information. All received data are processed through coordinate conversion and data verification to convert to the vehicle coordinate system with the vehicle center as the origin, and to provide a consistent environment model for subsequent path planning. The data verification includes range check and consistency check, such as whether the obstacle position coordinates are within the effective detection range of the sensor, whether the detection results of the same obstacle by different sensors are consistent, etc., to ensure the reliability and availability of the environment data.

[0062] In the process of judging whether the driving system is in or about to enter the performance degradation state according to the frequency domain characteristics coincidence degree of the historical load spectrum and the real-time operating condition, the historical load spectrum and the real-time operating condition are respectively subjected to frequency domain transformation to obtain corresponding historical frequency domain characteristics and real-time frequency domain characteristics. The frequency domain transformation is realized by using the fast Fourier transform method, and the historical load spectrum data and the real-time operating condition data in the time domain are converted into frequency domain representation.

[0063] The historical load spectrum data comes from previously stored drive system operation records, containing torque and speed sequences changing over time, where torque is in units of Newton-meters and speed is in units of revolutions per minute; real-time operation condition data includes currently collected drive motor torque command, actual output torque and speed values, and the collection method of these data is as described in the foregoing steps. Before frequency domain transformation, pre-processing operations are performed on the input data, including removing direct current components and abnormal value filtering, such as using a sliding average filter with a window length of 5 sampling points to smooth data fluctuations. During the transformation process, appropriate sampling frequency and window function are set, the sampling frequency is usually selected according to the characteristics of the drive system, for example, set to 1000 Hz, and the window function can be selected as the Hanning window to reduce spectral leakage. The historical frequency domain features and real-time frequency domain features obtained after frequency domain transformation contain amplitude spectrum and phase spectrum information, these features represent the intensity distribution of each frequency component in the form of complex numbers, the frequency resolution is determined by the sampling frequency and the number of sampling points, for example, when the sampling frequency is 1000 Hz and the number of sampling points is 1024, the frequency resolution is about 0.98 Hz.

[0064] Next, the distribution similarity of the historical frequency domain features and the real-time frequency domain features in the key frequency band is calculated as the frequency domain feature compliance. The key frequency band refers to the frequency range sensitive to the operating characteristics of the drive system, which is determined by analyzing the energy concentrated frequency band in the historical data, for example, the 50 Hz to 200 Hz frequency band, which is pre-marked according to the mechanical resonance characteristics and electromagnetic characteristics of the drive system. The distribution similarity calculation specifically includes extracting the power spectral density distribution of the historical and real-time operation condition signals in the main energy frequency band range, the power spectral density is obtained by squaring the amplitude spectrum of the frequency domain features and dividing by the frequency resolution. The power spectral density distribution correlation coefficient of the historical and real-time operation condition signals in the same key frequency band is calculated, the correlation coefficient is calculated using the Pearson correlation coefficient method, which is obtained by calculating the product of the covariance of the two power spectral density sequences and the product of their standard deviations, the calculation formula is: the correlation coefficient is equal to the covariance divided by the product of the standard deviations of the two sequences, the calculation result takes a value in the range of negative 1 to positive 1. The calculated power spectral density distribution correlation coefficient is used as a quantitative indicator of the frequency domain feature compliance, the closer the indicator value is to positive 1, the higher the similarity between the historical and real-time frequency domain features, and when the correlation coefficient is lower than the set threshold, it indicates that the operating characteristics of the system have changed significantly.

[0065] Finally, the state is judged based on the frequency domain feature coincidence degree. If the frequency domain feature coincidence degree is lower than the preset coincidence threshold, it is determined that the driving system is in or about to enter the performance degradation state. The preset coincidence threshold is determined by statistical analysis method, and is calculated based on a large number of historical data under normal state, for example, 1000 groups of frequency domain feature coincidence degree data under normal working conditions are collected, and the 5th percentile of the data distribution is taken as the threshold, the typical value range is between 0.7 and 0.9, and the specific value is adjusted according to the type of driving system and application scene. A hysteresis interval is set in the judgment process to avoid frequent state switching, for example, when the frequency domain feature coincidence degree is lower than the threshold 0.8, it is determined that the performance degradation state, and it needs to be restored to 0.85 or more to remove the state. The determination result is output as a binary state flag, and the determination time point and the corresponding frequency domain feature coincidence degree value are recorded, which provides decision basis for the subsequent thermal management strategy. The whole judgment process is executed at a fixed period, for example, once every 100 milliseconds, to ensure timely detection of driving system state change. During the execution process, an abnormal processing mechanism is also included, when the input data is abnormal or the calculation fails, the last valid determination result is maintained, and the fault code is recorded.

[0066] Through the above steps, the driving system performance degradation state judgment based on the frequency domain feature coincidence degree is realized, which provides accurate state input for the subsequent thermal management strategy. The frequency domain analysis method can effectively capture the change of driving system running characteristics, and timely discover the performance degradation trend, which provides guarantee for the safe operation of electric vehicles.

[0067] When in or about to enter the performance degradation state, the time domain trade-off evaluation is carried out based on the thermal inertia characteristics of the driving system and the urgency of the obstacle avoidance action. The process of generating the thermal management strategy includes the following steps. First, the thermal relaxation time of the driving system is estimated based on the thermal capacity and thermal resistance parameters of the driving system to represent the thermal inertia characteristics, wherein the thermal capacity parameter represents the heat absorbed per degree Celsius (J / ℃) required by the driving system to absorb per degree Celsius, and the thermal resistance parameter represents the temperature rise per watt (℃ / W) under unit heat flow. These parameters are obtained through the design specification of the driving system or through experimental test calibration. The experimental test can use the step heating method, that is, a constant power is applied to the driving system for heating, and the temperature change curve is recorded, and the thermal capacity and thermal resistance parameters are obtained by curve fitting. The thermal relaxation time is calculated by the product of the thermal capacity and the thermal resistance, and the unit is second (s). For example, when the thermal capacity is 200 J / ℃ and the thermal resistance is 0.5 ℃ / W, the thermal relaxation time is calculated as 100 s. This parameter reflects the response speed of the temperature change of the driving system, and the larger the value, the greater the thermal inertia of the system.

[0068] Next, the time window of obstacle avoidance action is calculated according to the relative motion relationship of dynamic obstacles in the vehicle operating environment data to represent the urgency of obstacle avoidance action. The vehicle operating environment data includes the position and speed information of surrounding obstacles, which are obtained by environmental perception sensors such as millimeter wave radar and laser radar. The time window of obstacle avoidance action is calculated by dividing the relative distance between the vehicle and the obstacle by the relative speed, with the unit in seconds (s), for example, when the relative distance is 10 m and the relative speed is 5 m / s, the time window of obstacle avoidance action is calculated as 2 s. In the calculation process, multiple obstacles are considered, and the smallest time window of obstacle avoidance action is selected as the final value, while a minimum time window threshold is set, for example, 0.5 s, to avoid unreasonable situations caused by too small calculation values. The threshold is determined according to the vehicle braking performance and the response time of the control system.

[0069] Then, the ratio of the thermal relaxation time to the time window of obstacle avoidance action is taken as the time domain trade-off factor. This calculation is a simple division operation, for example, when the thermal relaxation time is 100 s and the time window of obstacle avoidance action is 2 s, the time domain trade-off factor is calculated as 50. The time domain trade-off factor is a dimensionless value, and its size reflects the relative urgency of thermal management demand and obstacle avoidance demand. The larger the value, the more urgent the obstacle avoidance action. In the calculation process, a zero division protection mechanism is set, and when the time window of obstacle avoidance action approaches zero, the time domain trade-off factor is set to a maximum value, for example, 1000, to ensure the stability of the calculation process.

[0070] Finally, according to the continuous numerical interval where the time domain trade-off factor is located, the thermal management strategy guide that smoothly transitions between thermal protection and obstacle avoidance performance is dynamically generated. The value range of the time domain trade-off factor is divided into three continuous intervals, for example, 0 to 30 is the thermal protection priority interval, 30 to 70 is the balance interval, and 70 and above is the obstacle avoidance performance priority interval. In each interval, the linear interpolation method is used to calculate the degree of inclination of the thermal management strategy, for example, in the thermal protection priority interval, the thermal protection weight linearly decreases from 1.0 to 0.7, and the obstacle avoidance performance weight linearly increases from 0.0 to 0.3. The generated thermal management strategy guide includes target temperature limit values and power output limit values, which are calculated according to the weight coefficients, for example, the maximum allowed temperature decreases from 120℃ in the normal state to 100℃ in the thermal protection priority state, and the maximum output power decreases from 100kW in the normal state to 70kW in the thermal protection priority state. The entire evaluation process is executed at a fixed period, for example, the thermal management strategy guide is updated every 100ms, to ensure timely response to system state changes, while a first-order low-pass filter is used to smooth the strategy parameters to avoid unstable control system caused by parameter mutation.

[0071] The process of constructing the action feasibility space that satisfies both the internal dynamics constraints and the external space constraints according to the thermal management strategy orientation and the vehicle operating environment data, and obtaining the system performance constraint evaluation results by analyzing its dynamic envelope includes the following steps. First, map the thermal management strategy orientation to the constraint range of the vehicle longitudinal and lateral acceleration to form the internal dynamics constraints, where the thermal management strategy orientation includes the thermal protection weight coefficient and the obstacle avoidance performance weight coefficient, which are obtained through time domain trade-off evaluation. The specific implementation of time domain trade-off evaluation is described in the previous steps. The mapping process is implemented using linear interpolation method, for example, when the thermal protection weight coefficient is 0.8, the longitudinal acceleration limit range is adjusted from ±3.0 m / s² in normal state to ±1.5 m / s², and the lateral acceleration limit range is adjusted from ±2.0 m / s² in normal state to ±1.0 m / s². The calculation of acceleration constraint range is based on vehicle dynamics characteristics and driving system thermal state, which ensures to limit vehicle dynamic performance in thermal protection mode to avoid overheating. The interpolation coefficient is calibrated according to experimental data, for example, the acceleration limit value under different thermal states is determined through bench test.

[0072] Next, generate the geometric boundary of the vehicle drivable area according to the obstacle position, road boundary and traffic rules in the vehicle operating environment data to form the external space constraints. The obstacle position is obtained through environmental perception sensors, including the obstacle contour point cloud data detected by millimeter wave radar and laser radar, and the point cloud data is processed through clustering and classification to obtain obstacle boundary information. The road boundary is obtained by fusing lane line information obtained by visual sensor and high-precision map data, and the fusion algorithm adopts Kalman filtering method to associate and estimate the state of the visual detection result and the map data. Traffic rules include traffic restriction information such as speed limit signs and no overtaking areas, which are obtained through visual recognition and vehicle networking communication. Visual recognition uses deep learning algorithm to detect traffic signs, and vehicle networking communication uses DSRC or C-V2X technology to receive traffic control information. The geometric boundary of the drivable area is generated by using the polygon convex hull algorithm, which converts the obstacle position, road boundary and traffic rule restrictions into a series of continuous polygon vertex coordinates. These coordinates take the vehicle coordinate system as the reference system, with the origin at the vehicle center of mass, the X-axis pointing in the forward direction of the vehicle, and the Y-axis pointing to the left side of the vehicle.

[0073] A multi-constrained optimization problem is then formulated with inequality constraints of internal dynamics and equality constraints of external space. The internal dynamics constraints are represented as feasible regions in the vehicle state space, for example, the longitudinal acceleration constraint is represented as an inequality: -alongmax≤ along≤ alongmax, where along is the longitudinal acceleration and along max is the maximum allowed longitudinal acceleration value. The external space constraints are represented as spatial boundary conditions that the vehicle must obey, for example, the vehicle outline must not exceed the drivable region polygon boundary, which are represented as equalities: f(x, y) = 0, where x and y are the vehicle position coordinates and f is a function describing the drivable region boundary. The objective function of the optimization problem is set to minimize the curvature variation of the vehicle motion trajectory, and the decision variables include the vehicle position, velocity and acceleration. The mathematical form of the optimization problem is to find the vehicle state sequence that minimizes the objective function under the condition of satisfying all inequality constraints and equality constraints.

[0074] The action feasibility space is constructed by solving the multi-constrained optimization problem in real-time to obtain the feasible solution set. The current vehicle state initial values of the multi-constrained optimization problem are initialized at each control period, including the vehicle position, velocity, acceleration and heading angle, which are obtained by the vehicle state estimation module. A numerical optimization algorithm is used to solve the vehicle state set that satisfies the internal dynamics inequality constraints and the external space equality constraints, for example, using the interior point method or the sequential quadratic programming algorithm. The interior point method transforms the constrained optimization problem into an unconstrained optimization problem by introducing a barrier function, and the sequential quadratic programming algorithm approximates the optimal solution by iteratively solving quadratic programming sub-problems. The feasible solution set obtained by solving is used as the action feasibility space at the current time, which is represented as a multi-dimensional convex set in the vehicle state space, containing all possible vehicle states that satisfy the constraints. The dimension of the action feasibility space depends on the number of decision variables of the optimization problem, for example, when the decision variables are position and velocity, the action feasibility space is a four-dimensional space.

[0075] The analysis of the action feasibility space envelopes the coupling relationship between the curvature change of the envelope boundary and the volume shrinkage rate of the envelope in the prediction time domain. The prediction time domain is set to a time range of 3s to 5s, adjusted according to vehicle speed and environmental complexity, for example, set to 3s in urban road environment and 5s in highway environment. The curvature change of the envelope boundary is obtained by calculating the curvature change rate of the boundary of the action feasibility space at each point, with a unit of per meter (1 / m). The curvature is calculated using the differential geometry method, and the curvature value is obtained by calculating the first and second derivatives of the boundary curve. The volume shrinkage rate of the envelope is obtained by calculating the volume change rate of the action feasibility space over time, with a unit of cubic meters per second (m³ / s). The volume is calculated using numerical integration methods such as Monte Carlo integration or Simpson's integration. The coupling relationship analysis uses correlation analysis methods to calculate the correlation coefficient between the curvature change rate and the volume shrinkage rate. When the correlation coefficient exceeds 0.7, it is considered that there is a strong coupling relationship.

[0076] Based on the coordinated evolution mode of the curvature change of the envelope boundary and the volume shrinkage rate of the envelope, the system performance constraint evaluation result is determined. The time series data of the curvature change rate of the envelope boundary and the volume shrinkage rate of the envelope are monitored, with a sampling period of 100ms and a data length of 20 sampling points, corresponding to a time window of 2s. A time domain correlation model is established between the curvature change rate of the envelope boundary and the volume shrinkage rate of the envelope, and an autoregressive moving average model is used to describe the dynamic relationship between the two. The model parameters are estimated by the least squares method. When the coordinated deterioration mode of the curvature change intensifying and the volume accelerating shrinking is detected, the evaluation result of the intensification of the system performance constraint is generated. The judgment criteria of the coordinated deterioration mode include: the curvature change rate exceeds 0.5 per meter and the volume shrinkage rate exceeds 0.2 m³ / s, and the change trend of the two is consistent for more than 200ms. The evaluation result is output as a performance constraint level indicator, with a value range from 0 to 1. The larger the value, the more stringent the performance constraint. The performance constraint level indicator is calculated by a fuzzy logic system, with the curvature change rate and the volume shrinkage rate as inputs and the performance constraint level as output.

[0077] The entire construction and analysis process is executed in a fixed control period, usually 50ms to 100ms, to ensure that the action feasibility space can timely reflect the changes in the vehicle operating environment. All calculation processes contain an exception handling mechanism. When the optimization problem has no solution or the calculation times out, the feasible solution set of the last period is used as a backup scheme, and the system state is marked as needing manual intervention. In this way, real-time monitoring and performance evaluation of the vehicle operating state are realized, providing accurate input basis for subsequent path planning and dynamic obstacle avoidance decision-making.

[0078] The process of predicting the real-time maximum output capability boundary of the drive system according to the system performance constraint evaluation results includes the following steps. First, extract the envelope boundary curvature change and envelope volume shrinkage rate in the system performance constraint evaluation results, which are obtained by analyzing the dynamic envelope of the action feasibility space. The specific implementation is as described in the previous steps. The unit of envelope boundary curvature change is per meter (1 / m), which represents the degree of change in the shape of the boundary of the action feasibility space. The larger the value, the more dramatic the change in the boundary. The unit of envelope volume shrinkage rate is cubic meters per second (m³ / s), which represents the rate of reduction of the available range of the action feasibility space. A positive value indicates a contraction of the space, and a negative value indicates an expansion of the space. The extraction process reads the stored performance evaluation data through a data interface and verifies the validity of the data. The envelope volume shrinkage rate is usually between -5 m³ / s and 5 m³ / s. When the data exceeds this range, it is considered abnormal data and is processed accordingly.

[0079] Next, map the envelope boundary curvature change to the maximum torque limit of the drive system's sustainable output. The mapping relationship is established using a linear function combined with amplitude limiting processing. The mapping function is of the form: maximum torque limit = rated torque x (1 - k x envelope boundary curvature change), where k is the mapping coefficient, which is obtained by drive system characteristic experiment calibration.

[0080] The experimental calibration process includes testing the maximum sustainable output torque of the drive system under different curvature change conditions. The value of the mapping coefficient k is determined by data fitting, and the typical value range is 0.1 to 0.2 per meter. The unit of maximum torque limit is Newton-meter (N·m). The calculation process takes into account the thermal state and mechanical characteristics of the drive system to ensure that the output torque limit value is within the safe operating range of the drive system.

[0081] Then map the envelope volume shrinkage rate to the maximum allowable speed change rate limit of the drive system. The mapping process uses a lookup table combined with linear interpolation calculation to establish the correspondence between envelope volume shrinkage rate and maximum speed change rate limit. The mapping table contains multiple data points, such as when the envelope volume shrinkage rate is -5 m³ / s, the maximum speed change rate limit is set to 1500 r / min / s; when the envelope volume shrinkage rate is 0 m³ / s, the maximum speed change rate limit is set to 1000 r / min / s; when the envelope volume shrinkage rate is 5 m³ / s, the maximum speed change rate limit is set to 500 r / min / s. The parameters of the mapping table are determined by system testing, which includes measuring the maximum speed change rate that the drive system can withstand under different volume shrinkage rates. The inertia characteristics and control response characteristics of the drive system are considered during the testing process. The unit of maximum speed change rate limit is revolutions per minute per second (r / min / s). During the calculation process, linear interpolation is used to calculate the corresponding limit value based on the current envelope volume shrinkage rate.

[0082] Finally, the maximum torque limit and the maximum speed change rate limit are combined to generate the real-time maximum output capability boundary of the drive system in the prediction time domain. The method uses a multi-constraint optimization approach, taking the maximum torque limit and the maximum speed change rate limit as constraint conditions to calculate the feasible working area of the drive system in the prediction time domain. The prediction time domain is usually set to a time range of 3s to 5s, which is dynamically adjusted according to the vehicle operating state, for example, a longer prediction time domain is used in high-speed working conditions, and a shorter prediction time domain is used in urban working conditions. The real-time maximum output capability boundary is represented as a feasible region on the torque-speed plane, and the boundary of the region is determined by the maximum torque limit curve and the maximum speed change rate limit curve. The generation process includes sampling multiple working points on the torque-speed plane, verifying whether each working point satisfies the maximum torque limit and the maximum speed change rate limit, and setting the working point set that satisfies all constraints as the real-time maximum output capability boundary. The generated maximum output capability boundary data is transmitted to the control system through the vehicle bus for subsequent path planning and motion control. The entire prediction process is executed at a fixed period, for example, updated every 100ms, to ensure real-time reflection of the state changes of the drive system. The calculation process includes an exception handling mechanism, which uses conservative default values as output when the input data is abnormal, for example, using 70% of the rated torque as the maximum torque limit and 800r / min / s as the maximum speed change rate limit to ensure safe operation of the system.

[0083] The process of replanning the dynamic obstacle avoidance path according to the real-time maximum output capability boundary and the vehicle operating environment data and controlling the electric vehicle to perform obstacle avoidance operation includes the following steps. First, generate a speed profile that meets the instantaneous power limit of the drive system based on the real-time maximum output capability boundary, where the real-time maximum output capability boundary is obtained through the previous step and is represented as a feasible working area on the torque-speed plane, which is jointly defined by the maximum torque limit curve and the maximum speed change rate limit curve. The speed profile generation uses a quintic polynomial curve fitting method to plan a speed change curve that meets the drive system power limit in the prediction time domain, which is usually set to 3s to 5s, and the specific value is dynamically adjusted according to the vehicle speed and environmental complexity. The generation of the speed profile takes into account the dynamic response characteristics of the drive system to ensure that the acceleration change rate does not exceed the maximum value allowed by the system, for example, the acceleration change rate is limited to within 5m / s³, which is determined by the dynamic response test of the drive system. The generated speed profile contains time-speed sequence data, and the time interval is usually set to 0.1s, with the speed unit in meters per second (m / s). The speed profile must satisfy all the constraint conditions of the real-time maximum output capability boundary.

[0084] Next, the obstacle position in the vehicle operating environment data is combined with the road geometry information, and an obstacle-avoiding trajectory that satisfies the vehicle kinematics is planned under the speed profile constraint. The vehicle operating environment data includes obstacle position, speed, and size information obtained through environmental perception sensors, as well as road boundary and lane line information. The data acquisition method is described in the previous steps. The obstacle-avoiding trajectory planning adopts a sampling-based method, such as the Rapidly-exploring Random Tree algorithm, to generate a series of candidate trajectories under the speed profile constraint, and then select the optimal trajectory according to the obstacle-avoiding effect and comfort indicators. The trajectory planning satisfies the vehicle kinematics constraints, including the maximum steering angle limit and the minimum turning radius limit, such as a maximum steering angle limit of 30 degrees and a minimum turning radius limit of 5m, which are determined according to the vehicle chassis characteristics. The planned obstacle-avoiding trajectory contains position, speed, and acceleration information, discretely represented at a time interval of 0.1s, and the total length of the trajectory matches the prediction time domain.

[0085] The obstacle-avoiding trajectory is then converted into drive motor torque instructions and steering system angular displacement instructions. The conversion process uses a vehicle inverse dynamics model, in which the required driving torque is calculated based on the longitudinal acceleration information in the trajectory, and the required steering angular displacement is calculated based on the path curvature information in the trajectory. The calculation of the drive motor torque instruction is based on the vehicle longitudinal dynamics balance relationship, and the total required driving force is the sum of the acceleration resistance, air resistance, rolling resistance, and slope resistance; among them, the acceleration resistance is proportional to the vehicle mass and longitudinal acceleration, the air resistance is proportional to the air density, air resistance coefficient, vehicle frontal area, and the square of the running speed, the rolling resistance is proportional to the rolling resistance coefficient, vehicle mass, and gravitational acceleration, and the slope resistance is proportional to the vehicle mass, gravitational acceleration, and the sine value of the road slope angle; the calculated total required driving force is converted into the required torque instruction of the drive motor, combined with the effective rolling radius of the drive wheel and the total efficiency of the transmission system; the vehicle mass is obtained in real time through the on-board mass sensor or state estimator, the air resistance coefficient, frontal area, and rolling resistance coefficient are obtained through wind tunnel test and bench test calibration, the effective rolling radius of the tire is determined through vehicle design parameters, and the total efficiency of the transmission system is obtained through powertrain bench test calibration, and the road slope is obtained through inertial navigation unit or high-precision map data.

[0086] The calculation of the steering system angular displacement command is based on the Ackerman steering geometry principle, and the required steering angle is calculated according to the trajectory path curvature and the vehicle wheelbase, and the calculation formula is: the steering angle is equal to the inverse tangent function value of the product of the vehicle wheelbase and the trajectory curvature, wherein the vehicle wheelbase is a fixed design parameter, and the trajectory curvature is calculated from the geometric characteristics of the planned trajectory. The converted command signal is filtered to ensure the smoothness of the command change, for example, a first-order low-pass filter is used to smooth the command sequence, and the cutoff frequency of the filter is set to 10Hz according to the response bandwidth requirement of the control system, and the filter time constant is calculated according to the system sampling period and the smoothness requirement.

[0087] Finally, the corresponding instructions are sent to the drive motor controller and the steering system controller through the vehicle bus to perform the obstacle avoidance operation. The vehicle bus uses a controller area network bus or an Ethernet bus, and the communication protocol complies with the corresponding automotive bus standard, such as the CAN 2.0B protocol or the 100BASE-T1 Ethernet protocol. The instruction sending adopts a periodic manner, and the sending period is consistent with the control system period, which is usually 10ms. The sent instruction data includes the drive motor torque command value and the steering system angular displacement command value, as well as the corresponding time stamp and verification information, and the verification information uses a cyclic redundancy check code to ensure the integrity of data transmission. The instruction execution process includes a safety monitoring mechanism, which monitors the instruction execution in real time, and starts the corresponding fault handling program when an execution exception is detected, for example, an alarm is triggered when the actual torque deviation exceeds 10% of the instruction torque, and a correction control is performed when the steering angle tracking error exceeds 2 degrees. During the entire obstacle avoidance operation execution process, the vehicle state and environmental changes are continuously monitored, and the obstacle avoidance strategy is adjusted in a timely manner as needed to ensure the safety and effectiveness of the obstacle avoidance operation. All control instructions and state information are recorded in a non-volatile memory for subsequent analysis and optimization.

[0088] The calculations involved in the embodiments are all de-dimensioned to obtain numerical values, and the preset parameters and threshold values in the calculations are set by those skilled in the art according to actual conditions.

[0089] It should be noted that the present application can be deployed in the device itself to realize embedded application, or run on PC or other terminal with user interface, so as to meet various hardware environments and use requirements.

[0090] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center through wireless or wired transmission. The wired transmission includes optical fiber, twisted pair, coaxial cable, etc. The wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0091] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and module can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0092] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0093] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical modules, which can be located in one place or distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0094] In addition, each functional module in the various embodiments of the present application can be integrated in one processing module, or each module can exist physically independently, or two or more modules can be integrated in one module.

[0095] If the functions are implemented in the form of software functional modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0096] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0097] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An automatic driving path planning and dynamic obstacle avoidance method for electric vehicles, characterized in that, Comprise: S1, real-time acquisition of electric vehicle driving system historical load spectrum, real-time operating conditions and vehicle operating environment data, wherein the historical load spectrum is obtained by acquiring the pre-stored load spectrum data of the driving system in the historical running period; The load spectrum data is specifically a statistical distribution of a series of load states experienced by the driving system in the historical running period, which records the duration and frequency of the driving motor under different torque and speed combinations; S2, according to the frequency domain feature coincidence degree of historical load spectrum and real-time operating condition, judge whether the driving system is in or about to enter the performance attenuation state; S3, when in or about to enter the performance attenuation state, based on the thermal inertia characteristics of the driving system and the urgency of the obstacle avoidance action, the time domain weighting evaluation is carried out, and the heat management strategy direction is generated; S4, according to the heat management strategy direction and the vehicle operating environment data, the action feasibility space which satisfies the internal dynamics constraint and the external space constraint at the same time is constructed, and the system performance constraint evaluation result is obtained by analyzing its dynamic envelope; S5, according to the system performance constraint evaluation result, the real-time maximum output capability boundary of the driving system is predicted; S6, according to the real-time maximum output capability boundary and the vehicle operating environment data, the dynamic obstacle avoidance path is replanned, and the electric vehicle is controlled to execute obstacle avoidance operation.

2. The method of claim 1, wherein, Real-time acquisition of electric vehicle driving system real-time operating condition and vehicle operating environment data, comprising: Collecting the torque instruction, actual output torque and speed of the driving motor at the current time as the real-time operating condition; Receiving the position, speed and road geometry information of the surrounding obstacles output by the environment perception sensor as the vehicle operating environment data. 3.The method of claim 1, wherein, According to the frequency domain feature coincidence degree of historical load spectrum and real-time operating condition, judge whether the driving system is in or about to enter the performance attenuation state, comprising: The historical load spectrum and real-time operating condition are respectively subjected to frequency domain transformation to obtain the corresponding historical frequency domain feature and real-time frequency domain feature; The distribution similarity of historical frequency domain feature and real-time frequency domain feature in key frequency band is calculated as the frequency domain feature coincidence degree: the power spectrum density distribution of historical and real-time operating condition signals in the main energy frequency band range is extracted; the power spectrum density distribution correlation coefficient of historical and real-time operating condition signals in the same key frequency band is calculated; the power spectrum density distribution correlation coefficient is taken as the quantitative index of frequency domain feature coincidence degree; If the frequency domain feature coincidence degree is lower than the preset coincidence degree threshold, it is determined that the driving system is in or about to enter the performance attenuation state. 4.The method of claim 1, wherein, When in or about to enter the performance attenuation state, based on the thermal inertia characteristics of the driving system and the urgency of the obstacle avoidance action, the time domain weighting evaluation is carried out, and the heat management strategy direction is generated, comprising: Estimating the thermal relaxation time of the driving system based on its thermal capacity and thermal resistance parameters to represent the thermal inertia characteristics; According to the relative motion relationship of the dynamic obstacles in the vehicle operating environment data, the obstacle avoidance action time window is calculated to represent the urgency of the obstacle avoidance action; The ratio of thermal relaxation time to obstacle avoidance action time window is taken as the time domain weighting factor; According to the continuous numerical interval where the time domain weighting factor is located, the heat management strategy direction which smoothly transitions between thermal protection and obstacle avoidance performance is dynamically generated.

5. The method of claim 4, wherein, The smaller the time domain trade-off factor is, the more the thermal management strategy orientation tends to prioritize thermal protection, and the larger the time domain trade-off factor is, the more the thermal management strategy orientation tends to prioritize obstacle avoidance performance.

6. The method of claim 1, wherein, According to the thermal management strategy orientation and the vehicle operating environment data, an action feasibility space that meets both internal dynamics constraints and external space constraints is constructed, and a system performance constraint evaluation result is obtained by analyzing the dynamic envelope thereof, including: mapping the thermal management strategy orientation into a constraint range of vehicle longitudinal and lateral acceleration to form internal dynamics constraints; generating a geometric boundary of a vehicle drivable area according to obstacle positions, road boundaries and traffic rules in the vehicle operating environment data to form external space constraints; establishing a multi-constraint optimization problem with internal dynamics constraints as inequality constraints and external space constraints as equality constraints; constructing an action feasibility space that is updated in real time with vehicle motion by iteratively solving a feasible solution set of the multi-constraint optimization problem; analyzing the coupling relationship between envelope boundary curvature change and envelope volume shrinkage rate in the action feasibility space within a prediction time domain; determining the system performance constraint evaluation result based on the collaborative evolution mode of the envelope boundary curvature change and the envelope volume shrinkage rate.

7. The method of claim 6, wherein, By iteratively solving the feasible solution set of the multi-constraint optimization problem, an action feasibility space that is updated in real time with vehicle motion is constructed, including: initializing the current vehicle state initial value of the multi-constraint optimization problem at each control period; solving a vehicle state set that meets the internal dynamics inequality constraints and the external space equality constraints using a numerical optimization algorithm; and taking the feasible solution set obtained as the action feasibility space at the current time.

8. The method of claim 6, wherein, Based on the collaborative evolution mode of the envelope boundary curvature change and the envelope volume shrinkage rate, the system performance constraint evaluation result is determined, including: monitoring the time series data of the envelope boundary curvature change rate and the envelope volume shrinkage rate; establishing a time domain correlation model between the envelope boundary curvature change rate and the envelope volume shrinkage rate; and generating an evaluation result of the intensification of the system performance constraint when the collaborative deterioration mode of the intensification of the curvature change and the acceleration of the volume shrinkage is detected. 9.The automatic driving path planning and dynamic obstacle avoidance method for electric vehicles according to claim 1, wherein, According to the system performance constraint evaluation result, the real-time maximum output capability boundary of the drive system is predicted, including: extracting the envelope boundary curvature change and the envelope volume shrinkage rate from the system performance constraint evaluation result; mapping the envelope boundary curvature change into a maximum torque limit that the drive system can sustain output; mapping the envelope volume shrinkage rate into a maximum speed change rate limit that the drive system allows; combining the maximum torque limit and the maximum speed change rate limit to generate the real-time maximum output capability boundary of the drive system within the prediction time domain.

10. The method of claim 1, wherein, According to the real-time maximum output capability boundary and the vehicle operating environment data, a dynamic obstacle avoidance path is replanned, and the electric vehicle is controlled to perform obstacle avoidance operations, including: generating a speed profile that meets the instantaneous power limit of the drive system based on the real-time maximum output capability boundary; planning an obstacle avoidance trajectory that meets the kinematics of the vehicle under the constraint of the speed profile based on the obstacle positions and road geometry information in the vehicle operating environment data; converting the obstacle avoidance trajectory into drive motor torque instructions and steering system angular displacement instructions; sending corresponding instructions to the drive motor controller and the steering system controller through the vehicle bus to perform obstacle avoidance operations.

Images