Search

CN-121989987-A - Unmanned vehicle control algorithm with emergency obstacle avoidance function

CN121989987ACN 121989987 ACN121989987 ACN 121989987ACN-121989987-A

Abstract

The invention relates to the technical field of unmanned vehicle control, and discloses an unmanned vehicle control algorithm with an emergency obstacle avoidance function, which comprises the steps of obtaining the transverse deviation between the current gesture of a vehicle and an obstacle avoidance target track, and determining error flux according to the product of the first derivative of the transverse deviation and the curvature change rate of the obstacle avoidance target track; the method comprises the steps of converting error flux into gain adjustment factors based on a preset nonlinear mapping function, carrying out real-time weighting weakening on proportional gain and differential gain in a steering closed-loop control loop to enable a steering actuator to be in a controlled undergain state in an obstacle avoidance switching-in stage, and recovering feedback gain according to an energy recovery curve after the error flux falls back to a physical linear threshold.

Inventors

  • DAI SIDE

Assignees

  • 厦门戴尔乐科技股份有限公司

Dates

Publication Date
20260508
Application Date
20260122

Claims (10)

  1. 1. An unmanned vehicle control algorithm with an emergency obstacle avoidance function is characterized by comprising the following steps: step S101, acquiring the transverse deviation between the current transverse posture of the vehicle and the obstacle avoidance target track in real time, and determining error flux according to the product of the first derivative of the transverse deviation and the curvature change rate of the obstacle avoidance target track; Step S102, inputting the error flux into a preset nonlinear mapping function to determine a gain adjustment factor, wherein the nonlinear mapping function limits the gain adjustment factor and the absolute value of the error flux to show an inverse proportion mapping relation; step S103, real-time weighting weakening is carried out on the proportional gain and the differential gain in the steering closed-loop control loop by utilizing the gain adjustment factor, so that the action instruction change rate of the steering actuator is within the rated physical response envelope curve of the steering actuator, and the vehicle is in a controlled undergain state; Step S104, continuously monitoring the fluctuation amount of the error flux, and determining a preset energy recovery curve when the error flux is smaller than or equal to a preset physical linearity threshold; Step S105, the proportional gain and the differential gain are compared according to the slope of the energy recovery curve to recover from control cycle to control cycle, and the adjusted control quantity is output to drive the steering actuator to execute obstacle avoidance action.
  2. 2. The unmanned vehicle control algorithm with the emergency obstacle avoidance function according to claim 1, wherein the step S102 specifically includes calculating a product of a first derivative of the lateral deviation and a curvature change rate of the obstacle avoidance target trajectory in real time, determining an adjustment pressure value, mapping the adjustment pressure value to a gain adjustment factor by using a preset ramp function, wherein the gain adjustment factor monotonically decreases as an absolute value of the adjustment pressure value increases, and a value range of the gain adjustment factor is defined between 0.2 and 0.8.
  3. 3. The unmanned vehicle control algorithm with an emergency obstacle avoidance function according to claim 1, further comprising obtaining slip rate parameters of a wheel end of the vehicle in real time to determine a road adhesion coefficient, dynamically restricting a change slope of the gain adjustment factor based on the road adhesion coefficient, and reducing a rate of attenuation of the proportional gain and the differential gain when the road adhesion coefficient is smaller than a preset adhesion threshold.
  4. 4. The unmanned vehicle control algorithm with the emergency obstacle avoidance function according to claim 1, further comprising determining a safe side position of the obstacle avoidance target trajectory with respect to the obstacle, decomposing the lateral deviation into a local coordinate system corresponding to the safe side position to obtain a safe side lateral component and a dangerous side lateral component, and executing the feedback gain attenuation action in a direction away from the obstacle based on the safe side lateral component by a preset weight coefficient and based on the dangerous side lateral component by a normal gain coefficient larger than the weight coefficient in step S103.
  5. 5. The unmanned vehicle control algorithm with the emergency obstacle avoidance function according to claim 1, further comprising superimposing a perturbation signal of a preset frequency to a target control command of the steering actuator, collecting a position feedback signal of the steering actuator, extracting a response characteristic of the position feedback signal, which is the same frequency as the perturbation signal, and identifying a dynamic impedance of the steering actuator based on the response characteristic Wherein the dynamic impedance The formula is satisfied: , wherein, To superimpose the amplitude of the perturbation signal on the target control command, For the response amplitude of the position feedback signal, For phase difference of position feedback signal relative to perturbation signal, dynamic impedance is used And (3) adjusting the recovery step length of the energy recovery curve.
  6. 6. The unmanned vehicle control algorithm with an emergency obstacle avoidance function according to claim 1, further comprising calculating a deviation residual amount of the feedback gain before and after adjustment, mapping the deviation residual amount to a longitudinal load adjustment signal, adjusting a brake distribution ratio of a front axle and a rear axle of the vehicle based on the longitudinal load adjustment signal, and increasing a vertical load of a steering wheel of the vehicle.
  7. 7. The unmanned vehicle control algorithm with the emergency obstacle avoidance function according to claim 1, wherein the step S105 specifically includes monitoring a curvature direction switching point of the obstacle avoidance target trajectory, determining a phase compensation pulse according to a value of a current lateral deviation when the curvature direction switching point is reached, and injecting the phase compensation pulse into a steering closed-loop control circuit to offset lateral kinetic energy accumulated by the vehicle.
  8. 8. The unmanned vehicle control algorithm with the emergency obstacle avoidance function according to claim 1, further comprising smoothing the lateral deviation with a low-pass filter to filter out noise components above a preset cutoff frequency, the preset cutoff frequency satisfying one-half of the steering actuator sampling frequency in step S101.
  9. 9. The unmanned vehicle control algorithm with the emergency obstacle avoidance function according to claim 1, further comprising, after the step S105 is performed, extracting the driving current fluctuation amount of the steering actuator in real time, suspending the execution of the recovery process corresponding to the energy recovery curve when the driving current fluctuation amount is greater than a preset current threshold, and maintaining the proportional gain and the differential gain at the current levels until the driving current fluctuation amount is less than or equal to the preset current threshold.
  10. 10. The unmanned vehicle control algorithm with the emergency obstacle avoidance function according to claim 1, wherein the energy recovery curve satisfies a second order critical damping characteristic, and the recovery period of the energy recovery curve satisfies 1.5 to 2.5 times the time required for the steering actuator to complete the rated stroke.

Description

Unmanned vehicle control algorithm with emergency obstacle avoidance function Technical Field The invention belongs to the technical field of unmanned vehicle control, and particularly relates to an unmanned vehicle control algorithm with an emergency obstacle avoidance function. Background The current unmanned vehicle control system maintains the state variable to converge on the target track by using a feedback regulator, and the conventional working condition controller eliminates the track deviation by using a fixed control gain to drive a chassis executing mechanism to execute transverse and longitudinal movements. The utility model provides an automatic emergency obstacle avoidance system of unmanned vehicles and a control method thereof, which is disclosed in Chinese patent application No. CN114559923B, wherein a consensus behavior migration algorithm CBMA is introduced to carry out obstacle avoidance track planning in combination with circular clustering, the center of gravity is in a space track point static planning level, a path is selected through iterative evaluation fitness function, a high-speed obstacle avoidance working condition generates calculation delay, the algorithm does not touch control intensity and physical input mutation depth, an active physical weakening mechanism for feedback gain is lacked, the actuator faces nonlinear saturation risk in the face of transient large deviation, track fluctuation or convergence phase difference is generated after obstacle avoidance is completed, and dynamic alignment of the control intensity and hardware real-time bearing capacity is difficult to realize. Therefore, how to dynamically adjust the feedback gain according to the error flux to avoid the saturation risk of the actuator, and realize smooth convergence of the obstacle avoidance process on the premise of reducing the calculation load of the processor, becomes the technical problem to be solved by the invention. Disclosure of Invention The invention provides an unmanned vehicle control algorithm with an emergency obstacle avoidance function, which comprises the following steps: step S101, acquiring the transverse deviation between the current transverse posture of the vehicle and the obstacle avoidance target track in real time, and determining error flux according to the product of the first derivative of the transverse deviation and the curvature change rate of the obstacle avoidance target track; Step S102, inputting the error flux into a preset nonlinear mapping function to determine a gain adjustment factor, wherein the nonlinear mapping function limits the gain adjustment factor and the absolute value of the error flux to show an inverse proportion mapping relation; step S103, real-time weighting weakening is carried out on the proportional gain and the differential gain in the steering closed-loop control loop by utilizing the gain adjustment factor, so that the action instruction change rate of the steering actuator is within the rated physical response envelope curve of the steering actuator, and the vehicle is in a controlled undergain state; Step S104, continuously monitoring the fluctuation amount of the error flux, and determining a preset energy recovery curve when the error flux is smaller than or equal to a preset physical linearity threshold; Step S105, the proportional gain and the differential gain are compared according to the slope of the energy recovery curve to recover from control cycle to control cycle, and the adjusted control quantity is output to drive the steering actuator to execute obstacle avoidance action. Preferably, the step S102 specifically comprises the steps of calculating the product of the first derivative of the transverse deviation and the curvature change rate of the obstacle avoidance target track in real time to determine an adjustment pressure value, and mapping the adjustment pressure value into a gain adjustment factor by utilizing a preset slope function, wherein the gain adjustment factor monotonically decreases along with the increase of the absolute value of the adjustment pressure value, and the value range of the gain adjustment factor is limited to be between 0.2 and 0.8. Preferably, the method further comprises the steps of obtaining slip rate parameters of the wheel end of the vehicle in real time to determine road adhesion coefficients, and dynamically restraining the change slope of the gain adjustment factors based on the road adhesion coefficients, wherein the weakening rates of the proportional gain and the differential gain are reduced under the condition that the road adhesion coefficients are smaller than a preset adhesion threshold. Preferably, in step S103, the method further includes determining a safe side position of the obstacle avoidance target track relative to the obstacle, decomposing the lateral deviation into a local coordinate system corresponding to the safe side position to obtain a safe side lateral compone