US 12,280,809 B1
Human-machine shared control based on trajectory prediction
Panshuo Li, Guangzhou (CN); Kunsong Lin, Guangzhou (CN); Xinduan Yang, Guangzhou (CN); Hongyi Li, Guangzhou (CN); and Renquan Lu, Guangzhou (CN)
Assigned to Guangdong University of Technology, Guangzhou (CN)
Filed by Guangdong University of Technology, Guangzhou (CN)
Filed on May 10, 2024, as Appl. No. 18/660,261.
Claims priority of application No. 202311750441.9 (CN), filed on Dec. 18, 2023.
Int. Cl. B60W 60/00 (2020.01); B60W 40/10 (2012.01); B60W 50/00 (2006.01)
CPC B60W 60/0027 (2020.02) [B60W 40/10 (2013.01); B60W 50/0097 (2013.01); B60W 2050/0031 (2013.01); B60W 2520/06 (2013.01)] 7 Claims
OG exemplary drawing
 
1. A method for human-machine shared control based on trajectory prediction, characterized in that it comprises the following steps:
S1: establishing a road-vehicle model according to vehicle-related parameters;
S2: designing a path tracking controller according to the road-vehicle model: first, establishing a vehicle dynamics model and a vehicle kinematics model according to the vehicle-related parameters, and then establishing the road-vehicle model according to the vehicle dynamics model and the vehicle kinematics model;
wherein according to the vehicle dynamics model and the vehicle kinematics model, the road-vehicle model is obtained as follows:
x(t)=Ax(t)+Bu(t)±Bωω(t)
wherein

OG Complex Work Unit Math
wherein T is used as sampling time and Euler's method is utilized:

OG Complex Work Unit Math
converting a continuous vehicle model into a discrete vehicle model:
ξ(k+1)=Ãξ(k)+Bu(k)+Bωω(k)
wherein Ā=AT+I, B=BT, Bω=BωT, I denotes a unit matrix of corresponding dimensions;
wherein a prediction step of the path tracking controller is Np a control step is Nc, system measurements at each moment x(t) is used as a starting point of a prediction sequence x(0|t) to carry out optimization calculations, an optimal sequence of controlled variables is u*(k|t), k∈[0,Np−1] and a corresponding state sequences x*(k|t), k∈[1,Np] is obtained within prediction time domain [t,t+Np] and a first value of an optimal control sequence u*(0|t) is used as a control input at a current moment for acting on the vehicle;
wherein the path tracking controller based on a model predictive control method is designed with an objective function:
min J(x(k),u(k),ω(k))=min(YTQY+ΔUTRΔU)
s.t. ξ(k+1)=Ãξ(k)+Bu(k)+Bωω(k)
wherein, ξ(k+1)=Ãξ(k)+Bu(k)+Bωω(k) is the road-vehicle model, a predicted output of the system Y=[y(k+1), y(k+2), . . . , y(k+Np)]T, Q denotes an output weight matrix, a control increment sequenceΔU=[Δu(k),Δu(k+1), . . . ,Δu(k+Nc−1)]T, R denotes a control weight matrix, and in addition for achieving a goal of path tracking, an output equation is taken as

OG Complex Work Unit Math
considering limit constraints and increment constraints of controlled variables:
umin(t+k)≤u(t+k)≤umax(t+k)
Δumin(t+k)≤Δu(t+k)≤Δumax(t+k)
k=0,1, . . . ,Nc−1
S3: predicting driving trajectories based on the path tracking controller;
S4: evaluating driving risk of predicted driving trajectories;
S5: performing a switching of control between an automatic system and a driver based on the driving risk.