这段代码使用二次规划来解决MPC优化问题在每个时间步,并将系统的状态存储在x_history中用于绘图。所得到的图显示了系统状态随时间的演变,服从于MPC问题中指定的约束和目标。
% Define system parameters
A = [0.8, 0.2; 0.6, 0.4];
B = [0.2; 0.1];
C = [0.5, 0.5];
x0 = [0; 0];
% Define MPC parameters
horizon = 10;
Q = eye(2);
R = 0.1;
% Define optimization options
options = optimoptions('quadprog', 'Display', 'off');
% Initialize states and inputs
x = x0;
u = zeros(horizon, 1);
% Loop for a number of steps
for i = 1:100
% Formulate and solve optimization problem
H = 2 * blkdiag(Q, R);
f = zeros(2 * horizon, 1);
Aeq = [];
beq = [];
lb = [];
ub = [];
for j = 1:horizon
if j == 1
Aeq = [Aeq; C * x];
beq = [beq; x0];
else
Aeq = [Aeq; C * A^(j-1)];
beq = [beq; 0];
end
Aeq = [Aeq, zeros(size(Aeq, 1), horizon - j)];
lb = [lb; -inf];
ub = [ub; inf];
end
Aineq = [zeros(horizon, horizon), -eye(horizon)];
bineq = zeros(horizon, 1);
[z, fval, exitflag] = quadprog(H, f, Aineq, bineq, Aeq, beq, lb, ub, [], options);
if exitflag ~= 1
disp('Optimization failed');
break;
end
u(1) = z(1);
x = A * x + B * u(1);
% Shift inputs for next iteration
u = [u(2:end); 0];
% Store states for plotting
x_history(:, i) = x;
end
% Plot results
figure
hold on
plot(x_history(1, :), x_history(2, :), 'b');
xlabel('x1');
ylabel('x2');
title('Model Predictive Control');