Local path planning method for pure electric automatic driving commercial vehicle under park logistics scene

Abstract

A local path planning method for pure electric automatic driving commercial vehicles in a park logistics scenario, characterized in that it comprises the following steps: 1, constructing a load-center of mass offset model of the commercial vehicle; 2, calculating the lateral and longitudinal acceleration and turning radius limit of the commercial vehicle under the current load according to the vehicle kinematics model; 3, converting the Cartesian coordinate system of the vehicle and obstacles on the global reference path into the Frenet coordinate system; 4, dynamically sampling the global reference path in the lateral and longitudinal directions; 5, generating L(s) and s(t) planning functions according to the sampling state, and then generating a local path trajectory; 6, evaluating the feasibility of the local path trajectory, and taking the local path trajectory that meets the vehicle dynamics requirements as the feasible trajectory; 7, using a multi-objective cost function to evaluate the cost of the feasible trajectory to obtain the optimal trajectory; 8, issuing the optimal trajectory to the control module to control the commercial vehicle to travel. Effect: The optimal local path planning is realized.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, and in particular to a method for local path planning of pure electric autonomous commercial vehicles in a park logistics scenario. Background Technology

[0002] With the rapid development of autonomous driving technology, the application of autonomous driving in closed environments for commercial vehicles is becoming increasingly widespread. Closed park logistics scenarios provide excellent conditions for the use of autonomous commercial logistics vehicles, enabling 24 / 7 transportation, improving work efficiency, eliminating traffic safety risks associated with driver fatigue, and possessing economic value. However, the various delivery and receiving points within these park logistics scenarios present challenges for the route planning of autonomous logistics vehicles. These challenges include dynamic changes in worker and cargo stacking, unstructured T-shaped narrow warehouse entrances, unstructured herringbone ramp entry, and structured L-shaped steep road entry. Figure 2 As shown. Existing local path planning methods include artificial potential field method, fast random number, Lattice method, etc.

[0003] The shortcomings of existing technologies: Traditional local path planning algorithms cannot cope with the influence of vehicle parameters such as large range of total mass variation, large vehicle size, and large turning radius in real-world applications. In addition, the entry scenario in closed parks is complex, and there are local path planning problems under the constraints of real-world application scenarios such as changes in road parameters such as slope, road width and curvature, and target parking location. Summary of the Invention

[0004] This invention provides a local path planning method for pure electric autonomous commercial vehicles in a park logistics scenario, which achieves optimal local path planning and meets the requirements for local path planning of pure electric commercial vehicles in the park, especially considering the vehicle restrictions at each parking point.

[0005] To achieve the above objectives, this invention provides a local path planning method for pure electric autonomous commercial vehicles in a park logistics scenario, the key of which includes the following steps:

[0006] Step 1: Construct a load-center-of-gravity offset model for commercial vehicles;

[0007] Step 2: Based on the vehicle kinematic model, calculate the lateral acceleration limit, longitudinal acceleration limit, braking distance, and turning radius limit of the commercial vehicle under the current load.

[0008] Step 3: Perform global path planning based on the starting and ending coordinates of the commercial vehicle to obtain the global reference path. Convert the Cartesian coordinate system of the vehicles and obstacles on the global reference path to the Frenet coordinate system to reduce the planning complexity.

[0009] Step 4: Perform dynamic horizontal and vertical sampling of the global reference path in the Frenet coordinate system;

[0010] Step 5: Based on the sampling state, generate the L(s) planning function and the s(t) planning function, and perform trajectory fitting using the L(s) planning function and the s(t) planning function to generate at least two local path trajectories;

[0011] Step 6: Perform a feasibility assessment on each of the local path trajectories to determine whether the trajectory meets the vehicle dynamics requirements, screen the trajectories for feasibility, and select the local path trajectories that meet the vehicle dynamics requirements as feasible trajectories.

[0012] Step 7: Construct a multi-objective cost function, use the multi-objective cost function to evaluate the cost of the feasible trajectory, dynamically allocate the weights of the cost function, and then transform the optimal trajectory obtained from the Frenet coordinate system into the Cartesian coordinate system;

[0013] Step 8: Send the optimal trajectory in the Cartesian coordinate system to the control module, which then controls the commercial vehicle to travel along the optimal trajectory.

[0014] Through the above design, considering the characteristics of pure electric autonomous commercial vehicles and the road characteristics such as slope and curvature in logistics park scenarios, an improved Lattice algorithm is adopted. This algorithm takes into account the constraints of road slope, width, length, curvature, and parking space limitations, as well as the changes in vehicle size, turning radius, and total vehicle mass that cause changes in the center of gravity, thus affecting the steering and steering performance of commercial vehicles. This improves the local path planning to achieve optimal local path planning, which meets the local path planning requirements of pure electric commercial vehicles in the park, especially the vehicle constraints in each parking point scenario (slope, curve, narrow road).

[0015] Preferably, in step 1, when the commercial vehicle is unloaded, the position of the center of gravity is determined by the vehicle's own structure and obtained through design parameters or actual measurements.

[0016] r empty =(x empty ,y empty ,z empty )

[0017] Among them, (x empty ,y empty ,z empty ) represents the coordinates of the vehicle's unloaded center of mass, r empty This represents the unloaded centroid vector of the vehicle.

[0018] The mass m of each area of ​​the commercial vehicle is obtained through a pressure sensor array at the bottom of the commercial vehicle's cargo box. i(t), one pressure sensor collects the mass of a corresponding area, and a real-time calculation model for estimating the vehicle's center of gravity is established using the pressure sensor data:

[0019]

[0020] r total (r)=(x total ,y tota l,z total )

[0021] Where, m i (t) represents the mass of the i-th region at time t, m0 represents the unloaded mass of the vehicle, and r i (t) represents the centroid vector of the i-th region block, r total (r) is the centroid vector of the total mass of the vehicle after loading, (x) total ,y total ,z total The coordinates of the vehicle's center of gravity after loading, where n represents the total number of region blocks.

[0022] Preferably, in step 2, the effect of vehicle center of gravity shift caused by changes in vehicle load on vehicle dynamics is considered, where the center of gravity height z... total The lateral acceleration limit, which affects the roll threshold, is:

[0023] Based on dynamic constraints, the maximum lateral and longitudinal accelerations under various vehicle total mass conditions are calculated, and the expressions are as follows:

[0024]

[0025] Among them, a long,max a represents the maximum longitudinal acceleration. lat,max T represents the maximum lateral acceleration. max The maximum torque of the motor is represented by v, the vehicle speed by v, the total mass of the vehicle by m, and the tire rolling radius by r. t Indicates wheelbase, z total The Z-axis represents the distance between the vehicle's center of mass and the ground, given its total mass; g represents the acceleration due to gravity.

[0026] F resistance The total resistance, including air resistance, rolling resistance, and gradient resistance, is expressed as:

[0027]

[0028] Where ρ represents air density, C d The drag coefficient is represented by A, the frontal projected area of ​​the vehicle is represented by θ, and f is represented by f. r Indicates rolling resistance;

[0029] Lateral offset of the centroid ytotal The expression that causes changes in the vehicle's inherent steering characteristics and turning radius is as follows:

[0030]

[0031] Where, k max Indicates the maximum turning radius, L represents the wheelbase, and C represents the maximum turning radius. αf C represents the front wheel slip ratio. αr L represents the rear wheel slip ratio. r L represents the distance from the center of mass to the rear axle. f This indicates the distance from the center of mass to the front axle.

[0032] Change in centroid shift L r and L f This affects k max .

[0033] Preferably, in step 4, to address the impact of changes in the total mass of commercial vehicles on their steering, braking, and acceleration performance, as well as the path safety and feasibility of local path planning under changes in road curvature, slope, and width, the global reference path in the Frenet coordinate system is dynamically sampled longitudinally and laterally, as follows:

[0034] (1) Sampling in the longitudinal s direction:

[0035] The initial sampling interval Δs0 is based on the current velocity v and the time step T, and is expressed as:

[0036] Δs0=vT

[0037] The dynamic adjustment formula adjusts the sampling interval Δs by comprehensively considering the load mass influence factor α(m), the road curvature influence factor β(c), and the road slope influence factor γ(θ). The expression is as follows:

[0038]

[0039] Ensure Δs is within a reasonable range [Δs min ,Δs max ]Inside;

[0040] Generate a sequence of sampling points, incrementing the vertical sampling points s0, s1, s2, ... by Δs, and combine this with the horizontal offset to generate candidate local path trajectories;

[0041] (2) Sampling in the lateral d direction

[0042] Lane constraints: Lateral offset is limited within the lane d∈[d left d right ];

[0043] Curvature constraint: Lateral acceleration limitation to prevent sideslip.

[0044]

[0045] Among them, w vehicle Where μ is the vehicle width, μ is the road adhesion coefficient, and k is the road surface adhesion coefficient. road For the road curvature, d max For the maximum lateral displacement, d left d represents the distance from the center of the lane to the left lane line. right This indicates the distance from the center of the lane to the right lane line.

[0046] Preferably, the calculation expression for the load mass influence factor α(m) is as follows:

[0047]

[0048] Where m0 is the vehicle's unloaded mass; increasing the load mass m will lengthen the braking distance. When m > m0, the load mass influence factor α(m) increases, and the sampling interval needs to be reduced.

[0049] The calculation expression for the road curvature influence factor β(c) is as follows:

[0050] β(c)=1+K c |C|

[0051] Among them, K c The curvature sensitivity coefficient controls the intensity of curvature adjustment to the interval; C represents curvature, with units of rad / m. The larger the curvature C, the denser the sampling points.

[0052] The calculation expression for the road slope influence factor γ(θ) is as follows:

[0053] γ(θ)=1+K θ |sinθ|

[0054] Among them, K θ Here, θ is the slope sensitivity coefficient, and θ is the slope angle. The slope θ affects braking performance, and the braking distance increases when going downhill.

[0055] Preferably, in step 5, the L(s) programming function and the s(t) programming function are generated based on the sampling state, as follows:

[0056] In the Lattice algorithm, trajectories are generated based on the Frenet coordinate system, and the vehicle state is described by longitudinal (s) and lateral (d) components:

[0057] Vertical state:

[0058] Among them, s t Indicates position, Indicates speed, Indicates acceleration;

[0059] Lateral state:

[0060] Where, d s Indicates lateral offset, Indicates lateral offset speed, Indicates lateral offset acceleration;

[0061] (1) Generation of lateral trajectory

[0062] The relationship between the lateral displacement L(s) and the longitudinal displacement s, i.e., the programming function expression for L(s), is:

[0063] L(s) = a0 + a1s + a2s 2 +a3s 3 +a4s 4 +a5s 5

[0064] Where a0, a1, a2, a3, a4, and a5 are lateral displacement coefficients, which are solved by linear equations.

[0065] Boundary conditions:

[0066] Starting point constraint, s = 0:

[0067] L(0)=d current

[0068]

[0069] Where, d current This indicates the lateral offset from the starting point. This indicates the lateral offset velocity at the starting point. This indicates the lateral offset acceleration at the starting point;

[0070] Endpoint constraint, s = s T :

[0071] L(s T )=d target

[0072] L′(s T ) = 0

[0073] L″(s T ) = 0

[0074] Where, d target This indicates the lateral offset distance at the endpoint.

[0075] (2) Longitudinal velocity planning

[0076] The longitudinal displacement s(t) uses a fourth-order polynomial to satisfy the continuity of velocity and acceleration. The programming function expression for s(t) is:

[0077] s(t)=b0+b1t+b2t 2 +b3t 3 +b4t 4

[0078] Where b0, b1, b2, b3, and b4 are longitudinal displacement coefficients;

[0079] The planning starts at t=0 and ends at t=T:

[0080] Initial state:

[0081] s(0)=0

[0082] v(0)=v current

[0083] a(0)=a current

[0084] End point status:

[0085]

[0086] Among them, v current Let a represent the initial velocity. current v represents the initial acceleration. target This indicates the speed at the final destination.

[0087] In the Lattice algorithm, the lateral displacement function L(s) and the longitudinal displacement function s(t) are generated through polynomial parameterization. The generation of these two functions must satisfy the vehicle's dynamic constraints and boundary conditions, such as the starting and ending positions, velocities, and accelerations.

[0088] As a preferred option: In step 6, the feasibility of each local path trajectory is evaluated to determine whether the trajectory meets the vehicle dynamics requirements, and trajectory feasibility screening is performed. The process is as follows:

[0089] Longitudinal acceleration limit:

[0090] |a long |≤a long,max

[0091] Lateral acceleration limit:

[0092] |a lat |≤a lat,max

[0093] Steering curvature constraint:

[0094] κ(s)≤κ max

[0095] Obstacle collision detection:

[0096] The safe distance model (which expands with increasing vehicle cargo weight) is expressed as follows:

[0097] r safe =r base +fΔm

[0098] Where, r safe Indicates the safe distance, r base Δm represents the basic safety distance when unloaded, f represents the adjustment coefficient, Δm represents the vehicle's cargo weight, and Δm = m - m0.

[0099] The collision detection formula is as follows:

[0100]

[0101] ||P ego (t)-P obs (t)‖2>r safe

[0102] Among them, a long a represents longitudinal acceleration. long,max a represents the maximum longitudinal acceleration. lat a represents lateral acceleration. lat,max κ(s) represents the maximum lateral acceleration, and κ(s) represents the turning radius at time s. max P represents the maximum steering curvature. ego (t) represents the vehicle's position coordinates at time t, P obs (t) represents the obstacle's position coordinates at time t, and |||2 represents the L2 norm, [0, t horizon [] indicates the detection time range.

[0103] Preferably, in step 7, a weighted multi-objective cost function is designed for the commercial vehicle scenario to calculate the cost of each feasible path, considering the smoothness, safety, and following of the reference trajectory. The expression for the multi-objective cost function is as follows:

[0104] J = w1J smooth +w2J obstacle +w3J longi

[0105] Among them, w1, w2, and w3 are weighting coefficients, which need to be adjusted according to the scenario; J smooth To smooth out the cost, J obstacle For the cost of obstacles, J longi Cost of tracking reference trajectory;

[0106] (1) The smoothing cost J smooth The expression is:

[0107] J smooth =J lat_jerk +J long_jerk

[0108] Reduce sudden acceleration and sharp turns:

[0109] lateral acceleration J lat_jerk Minimize:

[0110]

[0111] Longitudinal acceleration constraint J long_jerk (The greater the total mass of the vehicle, the higher the cost of vibration):

[0112]

[0113] Where m0 is the vehicle's unloaded mass;

[0114] (2) The obstacle cost J obstacle The expression is:

[0115] J obstacle =J brake +J lat

[0116] Dynamic braking distance J brake constraint:

[0117]

[0118] d obs (s i ) represents the trajectory point s i Distance d to the nearest obstacle obs When the total mass m of the vehicle increases, the braking distance increases proportionally. increase;

[0119] Lateral stability J lat constraint:

[0120]

[0121] R(s i ) is at trajectory point s i Road curvature radius; v(s) i ) is at trajectory point s i Vehicle speed, μ is the road friction coefficient;

[0122] Permissible lateral acceleration threshold: The larger the total vehicle mass m, the weaker the vehicle's rollover resistance. The threshold is calculated as follows: reduce;

[0123] Actual lateral acceleration: derived from the radius of curvature of the trajectory R(s) i ) and vehicle speed v(s) i The calculation reflects the centrifugal force during turning;

[0124] (3) The cost of the tracking reference trajectory J longi The expression is:

[0125] J longi =J speed +J dis

[0126]

[0127] Among them, J speed J represents the cost of speed deviation. dis v represents the sum of the lateral deviations between the planned path and the reference path. ref v represents the vehicle's reference speed. evaluate L(s) represents the planned trajectory speed of the vehicle. i ) represents the trajectory point s in the Frenet coordinate system. i Lateral deviation from the reference trajectory.

[0128] The beneficial effects of this invention are:

[0129] 1. To address the impact of changes in the total mass of commercial vehicles on their steering, braking, and acceleration performance, and to ensure the safety and feasibility of local path planning under varying road curvature, gradient, and width, dynamic sampling optimization is performed in the longitudinal and lateral directions within the Frenet coordinate system. ① Based on the impact of load mass changes on steering and braking performance, the spacing of sampling points and the smoothness requirements of the path are dynamically adjusted; ② Changes in road curvature affect the vehicle's steering requirements and stability. In the Lattice algorithm, the density of sampling points is increased in road sections with greater curvature to more precisely control the vehicle's steering and path smoothness.

[0130] 2. For commercial vehicle scenarios, a weighted multi-objective cost function is designed to calculate the cost of each alternative path, taking into account the smoothness, safety, and following of the reference trajectory under the characteristics of changes in the total mass of commercial vehicles.

[0131] 3. The improved Lattice planner can adapt to changes in the total mass of commercial vehicles and environmental changes during actual use, improving trajectory safety in complex logistics scenarios in industrial parks, and can be extended to various complex scenarios such as ports and mines. Attached Figure Description

[0132] Figure 1 This is a schematic diagram of the process of the present invention;

[0133] Figure 2 (a) is a schematic diagram of an autonomous commercial vehicle operating in a logistics park in the embodiment;

[0134] Figure 2 (b) is a schematic diagram of the unstructured herringbone slope for entering the warehouse in the embodiment;

[0135] Figure 2 (c) is a schematic diagram of the unstructured T-shaped narrow warehouse opening in the embodiment;

[0136] Figure 2 (d) is a schematic diagram of the structured L-shaped road steep slope entry into the garage in the embodiment;

[0137] Figure 3 This is a schematic diagram of horizontal and vertical point sampling in the Frenet coordinate system in the embodiment. Detailed Implementation

[0138] The present invention will be further described in detail below with reference to the accompanying drawings and specific examples. The following embodiments or drawings are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0139] like Figure 1 As shown: A local path planning method for pure electric autonomous commercial vehicles in a park logistics scenario, including the following steps:

[0140] Step 1: Construct a load-center-of-gravity offset model for commercial vehicles;

[0141] Step 2: Based on the vehicle kinematic model, calculate the lateral acceleration limit, longitudinal acceleration limit, braking distance, and turning radius limit of the commercial vehicle under the current load.

[0142] Step 3: Perform global path planning based on the starting and ending coordinates of the commercial vehicle to obtain the global reference path. Convert the Cartesian coordinate system of the vehicles and obstacles on the global reference path to the Frenet coordinate system to reduce the planning complexity.

[0143] Step 4: Perform dynamic horizontal and vertical sampling of the global reference path in the Frenet coordinate system;

[0144] Step 5: Based on the sampling state, generate the L(s) planning function and the s(t) planning function, and perform trajectory fitting using the L(s) planning function and the s(t) planning function to generate at least two local path trajectories;

[0145] Step 6: Perform a feasibility assessment on each of the local path trajectories to determine whether the trajectory meets the vehicle dynamics requirements, screen the trajectories for feasibility, and select the local path trajectories that meet the vehicle dynamics requirements as feasible trajectories.

[0146] Step 7: Construct a multi-objective cost function, use the multi-objective cost function to evaluate the cost of the feasible trajectory, dynamically allocate the weights of the cost function, and then transform the optimal trajectory obtained from the Frenet coordinate system into the Cartesian coordinate system;

[0147] Step 8: Send the optimal trajectory in the Cartesian coordinate system to the control module, which then controls the commercial vehicle to travel along the optimal trajectory.

[0148] In step 1, when the commercial vehicle is unloaded, the position of its center of gravity is determined by the vehicle's own structure and is obtained through design parameters or actual measurements:

[0149] r empty =(x empty ,y empty ,z empty )

[0150] Among them, (x empty ,y empty ,z empty ) represents the coordinates of the vehicle's unloaded center of mass, r empty This represents the unloaded centroid vector of the vehicle.

[0151] The mass m of each area of ​​the commercial vehicle is obtained through a pressure sensor array at the bottom of the commercial vehicle's cargo box. i (t), one pressure sensor collects the mass of a corresponding area, and a real-time calculation model for estimating the vehicle's center of gravity is established using the pressure sensor data:

[0152]

[0153] r total (r)=(x total ,y tota l,z total )

[0154] Where, m i (t) represents the mass of the i-th region at time t, m0 represents the unloaded mass of the vehicle, and r i (t) represents the centroid vector of the i-th region block, r total (r) is the centroid vector of the total mass of the vehicle after loading, (x) total ,y total ,z total The coordinates of the vehicle's center of gravity after loading, where n represents the total number of region blocks.

[0155] In step 2, the impact of vehicle center of gravity shift caused by changes in vehicle load on vehicle dynamics is addressed, where the center of gravity height z... total The lateral acceleration limit, which affects the roll threshold, is:

[0156] Based on dynamic constraints, the maximum lateral and longitudinal accelerations under various vehicle total mass conditions are calculated, and the expressions are as follows:

[0157]

[0158] Among them, a long,max a represents the maximum longitudinal acceleration. lat,max T represents the maximum lateral acceleration. max The maximum torque of the motor is represented by v, the vehicle speed by v, the total mass of the vehicle by m, and the tire rolling radius by r. t Indicates wheelbase, z total The Z-axis represents the distance between the vehicle's center of mass and the ground, given its total mass; g represents the acceleration due to gravity.

[0159] F resistance The total resistance, including air resistance, rolling resistance, and gradient resistance, is expressed as:

[0160]

[0161] Where ρ represents air density, C d The drag coefficient is represented by A, the frontal projected area of ​​the vehicle is represented by θ, and f is represented by f. r Indicates rolling resistance;

[0162] Lateral offset of the centroid y total The expression that causes changes in the vehicle's inherent steering characteristics and turning radius is as follows:

[0163]

[0164] Where, k max Indicates the maximum turning radius, L represents the wheelbase, and C represents the maximum turning radius. αf C represents the front wheel slip ratio. αr L represents the rear wheel slip ratio. r L represents the distance from the center of mass to the rear axle. f This indicates the distance from the center of mass to the front axle.

[0165] Change in centroid shift L r and L f This affects k max .

[0166] In step 4, to address the impact of changes in the total mass of commercial vehicles on their steering, braking, and acceleration performance, and to ensure the safety and feasibility of local path planning under variations in road curvature, slope, and width, the global reference path in the Frenet coordinate system is dynamically sampled longitudinally and laterally. Figure 3 As shown, the details are as follows:

[0167] (1) Sampling in the longitudinal s direction:

[0168] The initial sampling interval Δs0 is based on the current velocity v and the time step T, and is expressed as:

[0169] Δs0=vT

[0170] The dynamic adjustment formula adjusts the sampling interval Δs by comprehensively considering the load mass influence factor α(m), the road curvature influence factor β(c), and the road slope influence factor γ(θ). The expression is as follows:

[0171]

[0172] Ensure Δs is within a reasonable range [Δs min ,Δs max ]Inside;

[0173] Generate a sequence of sampling points, incrementing the vertical sampling points s0, s1, s2, ... by Δs, and combine this with the horizontal offset to generate candidate local path trajectories;

[0174] (2) Sampling in the lateral d direction

[0175] Lane constraints: Lateral offset is limited within the lane d∈[d left d right ];

[0176] Curvature constraint: Lateral acceleration limitation to prevent sideslip.

[0177]

[0178] Among them, w vehicle Where μ is the vehicle width, μ is the road adhesion coefficient, and k is the road surface adhesion coefficient. road For the road curvature, d max For the maximum lateral displacement, d left d represents the distance from the center of the lane to the left lane line. right This indicates the distance from the center of the lane to the right lane line.

[0179] The calculation expression for the load mass influence factor α(m) is as follows:

[0180]

[0181] Where m0 is the vehicle's unloaded mass; increasing the load mass m will lengthen the braking distance. When m > m0, the load mass influence factor α(m) increases, and the sampling interval needs to be reduced.

[0182] The calculation expression for the road curvature influence factor β(c) is as follows:

[0183] β(c)=1+K c |C|

[0184] Among them, K c The curvature sensitivity coefficient controls the intensity of curvature adjustment to the interval; C represents curvature, with units of rad / m. The larger the curvature C, the denser the sampling points.

[0185] The calculation expression for the road slope influence factor γ(θ) is as follows:

[0186] γ(θ)=1+K θ |sinθ|

[0187] Among them, K θ Here, θ is the slope sensitivity coefficient, and θ is the slope angle. The slope θ affects braking performance, and the braking distance increases when going downhill.

[0188] In step 5, based on the sampled state, the L(s) programming function and the s(t) programming function are generated, as follows:

[0189] In the Lattice algorithm, trajectories are generated based on the Frenet coordinate system, and the vehicle state is described by longitudinal (s) and lateral (d) components:

[0190] Vertical state:

[0191] Among them, s t Indicates position, Indicates speed, Indicates acceleration;

[0192] Lateral state:

[0193] Where, d s Indicates lateral offset, Indicates lateral offset speed, Indicates lateral offset acceleration;

[0194] (1) Generation of lateral trajectory

[0195] The relationship between the lateral displacement L(s) and the longitudinal displacement s, i.e., the programming function expression for L(s), is:

[0196] L(s) = a0 + a1s + a2s 2 +a3s 3 +a4s 4 +a5s 5

[0197] Where a0, a1, a2, a3, a4, and a5 are lateral displacement coefficients;

[0198] Boundary conditions:

[0199] Starting point constraint, s = 0:

[0200] L(0)=d current

[0201]

[0202] Where, d current This indicates the lateral offset from the starting point. This indicates the lateral offset velocity at the starting point. This indicates the lateral offset acceleration at the starting point;

[0203] Endpoint constraint, s = s T :

[0204] L(s T )=d target

[0205] L′(s T ) = 0

[0206] L″(s T ) = 0

[0207] Where, d target This indicates the lateral offset distance at the endpoint.

[0208] (2) Longitudinal velocity planning

[0209] The longitudinal displacement s(t) uses a fourth-order polynomial to satisfy the continuity of velocity and acceleration. The programming function expression for s(t) is:

[0210] s(t)=b0+b1t+b2t 2 +b3t 3 +b4t 4

[0211] Where b0, b1, b2, b3, and b4 are longitudinal displacement coefficients;

[0212] The planning starts at t=0 and ends at t=T:

[0213] Initial state:

[0214] s(0)=0

[0215] v(0)=v current

[0216] a(0)=a current

[0217] End point status:

[0218]

[0219] Among them, v current Let a represent the initial velocity.current v represents the initial acceleration. target This indicates the speed at the final destination.

[0220] In the Lattice algorithm, the lateral displacement function L(s) and the longitudinal displacement function s(t) are generated through polynomial parameterization. The generation of these two functions must satisfy the vehicle's dynamic constraints and boundary conditions, such as the starting and ending positions, velocities, and accelerations.

[0221] In step 6, the feasibility of each local path trajectory is evaluated to determine whether the trajectory meets the vehicle dynamics requirements, and trajectory feasibility screening is performed. The process is as follows:

[0222] Longitudinal acceleration limit:

[0223] |a long |≤a long,max

[0224] Lateral acceleration limit:

[0225] |a lat |≤a lat,max

[0226] Steering curvature constraint:

[0227] κ(s)≤κ max

[0228] Obstacle collision detection:

[0229] The safe distance model (which expands with increasing vehicle cargo weight) is expressed as follows:

[0230] r safe =r base +fΔm

[0231] Where, r safe Indicates the safe distance, r base Δm represents the basic safety distance when unloaded, f represents the adjustment coefficient, Δm represents the vehicle's cargo weight, and Δm = m - m0.

[0232] The collision detection formula is as follows:

[0233]

[0234] ||P ego (t)-P obs (t)‖2>r safe

[0235] Among them, a long a represents longitudinal acceleration. long,max a represents the maximum longitudinal acceleration. lata represents lateral acceleration. lat,max κ(s) represents the maximum lateral acceleration, and κ(s) represents the turning radius at time s. max P represents the maximum steering curvature. ego (t) represents the vehicle's position coordinates at time t, P obs (t) represents the obstacle's position coordinates at time t, and |||2 represents the L2 norm, [0, t horizon [] indicates the detection time range.

[0236] In step 7, a weighted multi-objective cost function is designed for the commercial vehicle scenario to calculate the cost of each feasible path, taking into account the smoothness, safety, and following of the reference trajectory. The expression for the multi-objective cost function is as follows:

[0237] J = w1J smooth +w2J obstacle +w3J longi

[0238] Among them, w1, w2, and w3 are weighting coefficients, which need to be adjusted according to the scenario; J smooth To smooth out the cost, J obstacle For the cost of obstacles, J longi Cost of tracking reference trajectory;

[0239] (1) The smoothing cost J smooth The expression is:

[0240] J smooth =J lat_jerk +J long_jerk

[0241] Reduce sudden acceleration and sharp turns:

[0242] lateral acceleration J lat_jerk Minimize:

[0243]

[0244] Longitudinal acceleration constraint J long_jerk (The greater the total mass of the vehicle, the higher the cost of vibration):

[0245]

[0246] Where m0 is the vehicle's unloaded mass;

[0247] (2) The obstacle cost J obstacle The expression is:

[0248] J obstacle =J brake +J lat

[0249] Dynamic braking distance J brake constraint:

[0250]

[0251] d obs (s i ) represents the trajectory point s i Distance d to the nearest obstacle obs When the total mass m of the vehicle increases, the braking distance increases proportionally. increase;

[0252] Lateral stability J lat constraint:

[0253]

[0254] R(s i ) is at trajectory point s i Road curvature radius; v(s) i ) is at trajectory point s i Vehicle speed, μ is the road friction coefficient;

[0255] Permissible lateral acceleration threshold: The larger the total vehicle mass m, the weaker the vehicle's rollover resistance. The threshold is calculated as follows: reduce;

[0256] Actual lateral acceleration: derived from the radius of curvature of the trajectory R(s) i ) and vehicle speed v(s) i The calculation reflects the centrifugal force during turning;

[0257] (3) The cost of the tracking reference trajectory J longi The expression is:

[0258] J longi =J speed +J dis

[0259]

[0260] Among them, J speed J represents the cost of speed deviation. dis v represents the sum of the lateral deviations between the planned path and the reference path. ref v represents the vehicle's reference speed. evaluate L(s) represents the planned trajectory speed of the vehicle. i ) represents the trajectory point s in the Frenet coordinate system. i Lateral deviation from the reference trajectory.

[0261] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A local path planning method for pure electric autonomous commercial vehicles in a park logistics scenario, characterized in that, Includes the following steps: Step 1: Construct a load-center-of-gravity offset model for commercial vehicles; Step 2: Based on the vehicle kinematic model, calculate the lateral acceleration limit, longitudinal acceleration limit, braking distance, and turning radius limit of the commercial vehicle under the current load. Step 3: Perform global path planning based on the starting and ending coordinates of the commercial vehicle to obtain the global reference path, and convert the Cartesian coordinate system of the vehicles and obstacles on the global reference path to the Frenet coordinate system; Step 4: Perform dynamic horizontal and vertical sampling of the global reference path in the Frenet coordinate system; Step 5: Based on the sampling state, generate the L(s) planning function and the s(t) planning function, and perform trajectory fitting using the L(s) planning function and the s(t) planning function to generate at least two local path trajectories; Step 6: Perform a feasibility assessment on each of the local path trajectories to determine whether the trajectory meets the vehicle dynamics requirements, screen the trajectories for feasibility, and select the local path trajectories that meet the vehicle dynamics requirements as feasible trajectories. Step 7: Construct a multi-objective cost function, use the multi-objective cost function to evaluate the cost of the feasible trajectory, dynamically allocate the weights of the cost function, and then transform the optimal trajectory obtained from the Frenet coordinate system into the Cartesian coordinate system; Step 8: Send the optimal trajectory in the Cartesian coordinate system to the control module, which then controls the commercial vehicle to travel along the optimal trajectory; In step 5, based on the sampled state, the L(s) programming function and the s(t) programming function are generated, as follows: In the Lattice algorithm, trajectories are generated based on the Frenet coordinate system, and the vehicle state is described by longitudinal (s) and lateral (d) components: Vertical state: ; in, Indicates position, Indicates speed, Indicates acceleration; Lateral state: ; in, Indicates lateral offset, Indicates lateral offset speed, Indicates lateral offset acceleration; (1) Generation of lateral trajectory The relationship between the lateral displacement L(s) and the longitudinal displacement s, i.e., the programming function expression for L(s), is: ； in, , , , , , It is the lateral displacement coefficient; Boundary conditions: Starting point constraint, s=0: ； ； ； in, This indicates the lateral offset from the starting point. This indicates the lateral offset velocity at the starting point. This indicates the lateral offset acceleration at the starting point; Endpoint constraints : ； ； ； in, This indicates the lateral offset distance at the endpoint. (2) Longitudinal velocity planning The longitudinal displacement s(t) uses a fourth-order polynomial to satisfy the continuity of velocity and acceleration. The programming function expression for s(t) is: ； in, This is the longitudinal displacement coefficient; The planning starts at t=0 and ends at t=T: Initial state: ； ； ； End point status: ； ； in, This represents the initial velocity. This represents the acceleration in the initial state. This indicates the speed at the final destination.

2. The method for local path planning of pure electric autonomous commercial vehicles in a park logistics scenario according to claim 1, characterized in that: In step 1, when the commercial vehicle is unloaded, the position of its center of gravity is determined by the vehicle's own structure and is obtained through design parameters or actual measurements: ； in, Indicates the coordinates of the vehicle's unloaded center of mass. This represents the unloaded centroid vector of the vehicle. The mass of various areas of the commercial vehicle is obtained through a pressure sensor array at the bottom of the commercial vehicle's cargo box. A real-time calculation model for estimating the vehicle's center of gravity is established using pressure sensor data. ； ； in, Indicates the first Each region in time quality Indicates the vehicle's unloaded weight. Indicates the first The centroid vector of each region block, The center-of-mass vector of the vehicle after loading. The coordinates of the vehicle's center of gravity after loading, where n represents the total number of region blocks.

3. The method for local path planning of pure electric autonomous commercial vehicles in a park logistics scenario according to claim 1, characterized in that: In step 2, the impact of vehicle center of gravity shift caused by changes in vehicle load on vehicle dynamics is addressed, where the center of gravity height... The lateral acceleration limit, which affects the roll threshold, is: Based on dynamic constraints, the maximum lateral and longitudinal accelerations under various vehicle total mass conditions are calculated, and the expressions are as follows: ； in, Indicates the maximum longitudinal acceleration. Indicates the maximum lateral acceleration. This indicates the maximum torque of the motor. Indicates vehicle speed; It represents the total mass of the vehicle, which is the total mass of the vehicle (unloaded) plus the mass of the cargo. Indicates the tire's rolling radius. Indicates wheelbase. This represents the distance between the vehicle's center of gravity and the ground along the Z-axis, given its total mass. Represents gravitational acceleration; The total resistance, including air resistance, rolling resistance, and gradient resistance, is expressed as: ； in, Indicates air density, Indicates the drag coefficient. Indicates the projected area of ​​the front of the vehicle. Indicates slope, Indicates rolling resistance; Lateral shift of the centroid The expression that causes changes in the vehicle's inherent steering characteristics and turning radius is as follows: ； in, Indicates the maximum turning radius. Indicates wheelbase. Indicates the front wheel slip ratio. Indicates the rear wheel slip ratio. This represents the distance from the center of mass to the rear axle. This indicates the distance from the center of mass to the front axle.

4. The method for local path planning of pure electric autonomous commercial vehicles in a park logistics scenario according to claim 1, characterized in that: In step 4, the global reference path is dynamically sampled horizontally and vertically in the Frenet coordinate system, as follows: (1) Sampling in the longitudinal s direction: Initial sampling interval Based on the current velocity v and time step T, the expression is: ； Dynamically adjust the formula to incorporate the load-mass influence factor. Road curvature influencing factors and road slope influencing factors Adjust sampling interval The expression is as follows: ； Ensure ∆s is within a reasonable range Inside; Generate a sequence of sampling points, and generate vertical sampling points by incrementing ∆s. , , …and combine with lateral offset to generate candidate local path trajectories; (2) Sampling in the lateral d direction Lane constraints: Lateral offset is limited to within the lane. ; Curvature constraint: Lateral acceleration limit: ； in, For vehicle width, The road adhesion coefficient, For road curvature, This represents the maximum lateral displacement. Indicates the distance from the center of the lane to the left lane line. This indicates the distance from the center of the lane to the right lane line.

5. The method for local path planning of pure electric autonomous commercial vehicles in a park logistics scenario according to claim 4, characterized in that: The load mass influence factor The calculation expression is as follows: ； in, Vehicle unloaded weight; The road curvature influencing factor The calculation expression is as follows: ； in, The curvature sensitivity coefficient; Indicates curvature, measured in rad / m; The road slope influencing factors The calculation expression is as follows: ； in, This is the slope sensitivity coefficient. The slope angle is denoted by .

6. The method for local path planning of pure electric autonomous commercial vehicles in a park logistics scenario according to claim 1, characterized in that: In step 6, the feasibility of each local path trajectory is evaluated to determine whether the trajectory meets the vehicle dynamics requirements, and trajectory feasibility screening is performed. The process is as follows: Longitudinal acceleration limit: ； Lateral acceleration limit: ； Steering curvature constraint: ； Obstacle collision detection: The safe distance model expression is as follows: ； in, Indicates a safe distance. Indicates the basic safety distance when unloaded. Indicates the adjustment factor. Indicates the vehicle's cargo capacity. ; The collision detection formula is as follows: ； ； in, Indicates longitudinal acceleration. Indicates the maximum longitudinal acceleration. Indicates lateral acceleration. Indicates the maximum lateral acceleration. This represents the turning radius at time s. Indicates the maximum steering curvature. This represents the vehicle's position coordinates at time t. This represents the coordinates of the obstacle's position at time t. Represents the L2 norm. Indicates the detection time range.

7. The method for local path planning of pure electric autonomous commercial vehicles in a park logistics scenario according to claim 1, characterized in that: In step 7, a weighted multi-objective cost function is designed for the commercial vehicle scenario to calculate the cost of each feasible path, taking into account the smoothness, safety, and following of the reference trajectory. The expression for the multi-objective cost function is as follows: ； in, These are the weighting coefficients. To smooth out the cost, As a result of obstacles, Cost of tracking reference trajectory; (1) The smoothing cost The expression is: ； Reduce sudden acceleration and sharp turns: lateral acceleration Minimize: ； Longitudinal acceleration constraint : ； in, The unloaded weight of the vehicle; (2) Cost of the obstacle The expression is: ； Dynamic braking distance constraint: ； Represents trajectory points Distance to the nearest obstacle When the total mass m of the vehicle increases, the braking distance increases proportionally. increase; Lateral stability constraint: ； For the trajectory point The radius of curvature of the road; For the trajectory point vehicle speed, Road friction coefficient; Permissible lateral acceleration threshold: The larger the total vehicle mass m, the weaker the vehicle's rollover resistance. The threshold is calculated as follows: reduce; Actual lateral acceleration: determined by the radius of curvature of the trajectory and vehicle speed v Calculations reflect the centrifugal force during turning; (3) The cost of the tracking reference trajectory The expression is: ； ； ； in, This represents the cost of speed deviation. This represents the sum of the lateral deviations between the planned path and the reference path. Indicates the vehicle's reference speed. This indicates the planned trajectory speed of the vehicle. Represents the trajectory points in the Frenet coordinate system Lateral deviation from the reference trajectory.

Images