CN-121989942-A - Intelligent control method and system for automatic driving electric hydrogen storage and charging robot

CN121989942ACN 121989942 ACN121989942 ACN 121989942ACN-121989942-A

Abstract

The invention provides an intelligent control method and system for an automatic driving electro-hydrogen storage and charging robot, and belongs to the technical field of intelligent control; the core of the invention is to construct a dynamic energy state potential field in real time, and the potential field quantifies the comprehensive energy guarantee capability of the robot for completing tasks under different time slots. Based on the state potential field and gradient information thereof, uniformly modeling the energy distribution of the hydrogen fuel cell and the power cell and the motion control of the robot into a multi-constrained collaborative optimization problem, and adopting an optimization theory based on a Hamiltonian to solve in real time so as to output a globally optimal driving and energy control instruction; the corresponding system comprises a state potential field construction, collaborative optimization, energy management and motion control module; the invention realizes the prospective deep coupling of the energy state and the movement behavior, and remarkably improves the task reliability, the energy economy and the overall operation efficiency of the robot in the dynamic environment.

Inventors

WEI FEI
CHEN PING
DUAN XIAOCHEN

Assignees

四川新工绿氢科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260330

Claims (10)

1. An intelligent control method of an automatic driving electric hydrogen storage and charging robot is characterized by comprising the following steps: Step S1, constructing a dynamic energy situation field E (p, t) in real time, wherein p represents the spatial position of a robot, and t represents time, and the dynamic energy situation field E (p, t) is a scalar field for representing the comprehensive energy guarantee capability of the current position of the robot for completing subsequent tasks at the moment t; step S2, calculating situation field gradient information of the current moment t based on the dynamic energy situation field E (p, t) E (p, t) and rate of change information E(p, t)/ t; Step S3, the situation field gradient information is processed E (p, t) and rate of change information E(p, t)/ T is used as a constraint condition, and a cooperative optimization problem of robot motion control and multi-source energy distribution is established; And step S4, solving the collaborative optimization problem to obtain an optimal control instruction at the current moment, wherein the control instruction at least comprises a torque instruction of a driving motor, a steering instruction, a power output instruction of a fuel cell and a charge and discharge power instruction of a power cell.
2. The intelligent control method of an automatic driving electro-hydrogen storage and charging robot according to claim 1, wherein the constructing the dynamic energy situation field E (p, t) in real time in step S1 specifically includes: Step S11, calculating a basic energy situation value E_base (p, t) according to real-time state data of the robot, wherein the basic energy situation value E_base (p, t) is determined at least based on the residual available energy of the robot and the minimum path energy consumption required by the current position to reach a task target point; Step S12, acquiring environment dynamic information, and calculating a situation correction term E_mod (p, t), wherein the situation correction term E_mod (p, t) corrects a basic energy situation value at least based on environment obstacle distribution, energy charging station queuing state and traffic flow information; And step S13, synthesizing the basic energy situation value E_base (p, t) and the situation correction term E_mod (p, t) to obtain a dynamic energy situation field E (p, t) =f (E_base (p, t), E_mod (p, t)) which is finally used for control.
3. The intelligent control method of an automatic driving electro-hydrogen storage and charging robot according to claim 2, wherein the calculating the basic energy situation value e_base (p, t) in step S11 specifically includes: E_base(p, t) = α * SOC(t) + β * H 2 _mass(t) - γ * min_path_energy(p, Goal); Wherein, SOC (t) is the state of charge of the power battery at the current moment, H 2 _mass (t) is the residual hydrogen mass of the hydrogen storage tank at the current moment, min_path_energy (p, goal) is the theoretical minimum path energy consumption from the current position p to the task target point Goal, and alpha, beta and gamma are weight coefficients determined according to the efficiency characteristics of the fuel battery and the power battery.
4. The intelligent control method of an automatic driving electro-hydrogen storage and charging robot according to claim 1, wherein the objective function J (u (t)) for constructing the collaborative optimization problem in step S3 is: J(u(t)) = ∫[w 1 * (T_task - t) + w 2 * P_total(t) + w 3 * ||E(p(t), t) - E_ref||²] dt, Wherein u (T) is a control input vector to be optimized, T_task is task cut-off time, P_total (T) is total power consumption of the system, E_ref is a desired energy situation reference value, and w 1 、w 2 、w 3 is a weight coefficient; Constraints on the optimization problem include: Constraint condition C1, namely kinematic constraint, |a (t) | is less than or equal to a_max, v (t) | is less than or equal to v_max (p), wherein a (t) is an instantaneous acceleration vector of the robot at a moment t, a_max is a maximum allowable acceleration scalar value of the robot, v (t) is an instantaneous speed scalar of the robot at the moment t, and v_max (p) is a local maximum allowable speed at a spatial position p; Constraint condition C2, namely dynamic energy situation constraint, wherein E (p (t+Δt), t+Δt) is not less than E_min, E (p (t+Δt), t+Δt) is a dynamic energy situation field value which is predicted to be possessed by the robot when the robot reaches a position p (t+Δt) at the next moment t+Δt, E_min is a safety threshold of the energy situation; Constraint condition C3, namely hydrogen fuel cell power climbing rate constraint, |dP_fc (t)/dt| is less than or equal to R_fc_max, wherein dP_fc (t)/dt is the output power change rate of the hydrogen fuel cell at time t, and R_fc_max is the maximum allowable climbing rate of the fuel cell power; Constraint condition C4, namely the charge and discharge multiplying power constraint of the power battery, wherein I_bat (t) I is less than or equal to I_bat_max, wherein I_bat (t) is the instantaneous current of the power battery at the moment t, and I_bat_max is the maximum continuous charge/discharge current absolute value allowed by the power battery.
5. The intelligent control method of an automatic driving electro-hydrogen storage and charging robot according to claim 1, wherein the method according to claim 1 is characterized in that the solving the collaborative optimization problem in step S4 adopts a hamilton minimization principle based on constraint expansion, and specifically comprises the following steps: Constructing a hamiltonian H (x, u, λ, μ, t) =l (x, u, t) +λ * f(x, u, t) + μ * C(x, u, t), By solving a system of equations H/ u = 0, = - H/ x, μ ≥ 0, μ * C=0, obtaining an optimal control command u (t) at the current moment; Wherein x is a system state vector, including position, speed, SOC, and hydrogen quality, L is an integrated function of the objective function in step S3, f is a system state equation, C is a vector composed of all constraint conditions, λ and μ are lagrangian multiplier vectors, u is a control input vector, and t is a time variable.
6. An intelligent control system of an automatic driving electric hydrogen storage and charging robot for realizing the intelligent control method of the automatic driving electric hydrogen storage and charging robot according to any one of claims 1-5, characterized by comprising: the state potential field construction module is used for constructing and updating a dynamic energy state field E (p, t) according to the real-time state and environmental information of the robot; the collaborative optimization module is connected with the situation field construction module and is used for constructing and solving a collaborative optimization problem of motion control and multi-source energy distribution based on the dynamic energy situation field and gradient information thereof to generate an optimal control instruction; The multi-source energy management module is connected with the collaborative optimization module and is used for accurately allocating the power output of the hydrogen fuel cell and the power cell according to the energy allocation instruction in the optimal control instruction; And the vehicle motion control module is connected with the collaborative optimization module and is used for driving a driving motor and a steering executing mechanism of the robot according to the motion control instruction in the optimal control instruction.
7. The intelligent control system of an autonomous electro-hydrogen storage robot of claim 6, wherein the situational field construction module comprises: The basic situation calculation unit is used for calculating a basic energy situation value E_base (p, t) of the current position of the robot according to the real-time energy state of the robot and a preset path planning algorithm; The environment sensing and correcting unit is used for acquiring environment dynamic information through the laser radar, the camera and the V2X communication equipment and calculating a correction term E_mod (p, t) for the basic energy situation value; and the situation fusion unit is used for fusing the basic energy situation value and the correction term and outputting a complete dynamic energy situation field E (p, t) and gradient information thereof.
8. The intelligent control system of an autonomous electro-hydrogen storage robot of claim 6, wherein the collaborative optimization module is built with an optimization solver configured to solve in real time using a hamilton minimization principle based on constraint expansion.
9. The intelligent control system of an autonomous electro-hydrogen storage robot of claim 6, wherein the multi-source energy management module comprises: A fuel cell controller for receiving the power output command and controlling the output power and efficiency of the fuel cell stack by adjusting the hydrogen supply amount and the air intake amount; The battery management system is used for receiving the charge and discharge power instruction and controlling the charge and discharge current and the working state of the power battery; And the power coupling unit is used for coupling the direct current output by the fuel cell with the charge and discharge power of the power cell and smoothly supplying the direct current to the driving motor and the vehicle-mounted auxiliary parts.
10. The intelligent control system of an autonomous electro-hydrogen storage robot of claim 6, further comprising a safety monitoring and switching module that monitors the value of the dynamic energy profile E (p, t) in real time, the safety monitoring and switching module switching control from the co-optimization module to a preset emergency energy conservation controller that enforces a conservation strategy of slowing down and driving to the nearest charging station when E (p, t) is below a preset safety threshold e_safe.

Description

Intelligent control method and system for automatic driving electric hydrogen storage and charging robot Technical Field The invention relates to the technical field of intelligent control, in particular to an intelligent control method and system for an automatic driving electro-hydrogen storage and charging robot. Background With the rapid development of new energy technology, the mobile storage and charging robot with the electro-hydrogen hybrid power system has the flexibility of pure electric drive and the convenience of rapid hydrogen energy supplement, and has great application potential in the scenes of regional energy distribution, emergency power supply, cooperative energy supplement of motorcades and the like. Such robots typically integrate lithium ion power batteries, hydrogen fuel cells, and an autopilot chassis, and their core tasks are to efficiently and safely accomplish material transfer, vehicle charging, or hydrogen addition instructions. Currently, control architectures for such complex systems commonly employ layered or modular designs. At the motion control level, a Model Predictive Control (MPC) or proportional-integral-derivative (PID) control algorithm based on path tracking is typically employed to ensure that the vehicle follows a preset trajectory accurately. At the energy management level, algorithms based on rules (e.g., state machines) or based on optimizations (e.g., equivalent fuel consumption minimization strategies) are often employed to distribute the output power between the hydrogen fuel cell and the power cell. And the path planning module independently calculates a running route with optimal time or distance according to the task target, the map information and the real-time perceived obstacle. However, the above-described conventional method of decoupling motion control, energy management and path planning has revealed significant technical drawbacks in practical applications, mainly in the following aspects: First, the energy state is fractured from the dynamic path environment, resulting in shortsightedness. The path planning in the existing scheme usually takes the shortest time or the shortest distance as a single target, and does not take the real-time energy state (such as the battery state of charge (SOC) and the residual hydrogen amount) of the robot and the time-varying characteristics thereof as core constraint of the planning. The planned path may include an implicit high-energy-consumption road section such as long-distance uphill, frequent start-stop, or high-speed driving. The independent energy management module only performs power distribution optimization on the power system level, and cannot foresee and actively cope with severe load changes caused by paths. Such a split results in the robot, when executing the "optimal" path, possibly running the risk of mid-way energy exhaustion, or being forced to enable a conservative global energy margin strategy, thus sacrificing task efficiency substantially. Second, constraint processing is static and cannot accommodate dynamic uncertainty. Existing control methods typically consider the physical constraints (e.g., maximum speed, acceleration) and the energy constraints (e.g., minimum SOC) of the robot as fixed thresholds. However, in a real operating environment, the availability of the charging stations, queuing time, traffic flow conditions, and ambient temperature (which severely affects fuel cell efficiency and internal cell resistance) are all dynamically changing. The static constraint model cannot integrate the real-time information, so that the control system lacks the capability of online re-planning and self-adaptive adjustment when encountering emergency (such as occupied target charging station) or environmental interference, and the robustness is insufficient. Third, multi-objective optimizations conflict with each other, lacking a global collaborative framework. Mission time, system energy efficiency, equipment life (e.g., to avoid frequent start-up and shut-down of the fuel cell and high rate discharge of the cell) and operational safety are a number of competing goals. Existing methods often reconcile these conflicts by presetting fixed weights or compromising them at subsequent control layers, lacking a unified mathematical framework that can dynamically trade off the above objectives according to real-time situations (e.g., remaining energy, task urgency). This tends to result in local optimization and global suboptimal, e.g., excessive energy consumption for catch-up time, rendering subsequent tasks unexecutable. Therefore, an innovative intelligent control method is urgently needed to solve the technical problems of energy management and motion planning disjoint, constraint model rigidification and multi-objective coordination difficulty existing in the control strategy of the automatic driving electric hydrogen storage and charging robot in the prior art. Disclosure of Invention The in