EmaMaker
/
thesis
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

switch to ub, lb constraints 

since reformulating the constraints A and b are no longer needed, this is easier
diffdrive-costfun
EmaMaker 2024-09-14 13:31:44 +02:00
parent 2408bdf472
commit 6dd6af2b8c
1 changed files with 9 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
control_act.m
View File

@ -30,19 +30,16 @@ function [u_corr, U_corr_history, q_pred] = ucorr(t, q, sim_data)
if eq(pred_hor, 0) if eq(pred_hor, 0)
return return
elseif eq(pred_hor, 1) elseif eq(pred_hor, 1)
H = eye(2); H = eye(2)*2;
f = zeros(2,1); f = zeros(2,1);
T_inv = decouple_matrix(q_act, sim_data); T_inv = decouple_matrix(q_act, sim_data);
ut = utrack(t, q_act, sim_data); ut = utrack(t, q_act, sim_data);
%A = [T_inv; -T_inv];
A = [eye(2); -eye(2)];
d = T_inv*ut; d = T_inv*ut;
b = [s_-d;s_+d];
% solve qp problem % solve qp problem
options = optimoptions('quadprog', 'Display', 'off'); options = optimoptions('quadprog', 'Display', 'off');
u_corr = quadprog(H, f, A, b, [],[],[],[],[],options); u_corr = quadprog(H, f, [], [], [],[], -s_ - d, s_-d, [], options);
q_pred = q_act; q_pred = q_act;
U_corr_history(:,:,1) = u_corr; U_corr_history(:,:,1) = u_corr;
@ -113,34 +110,25 @@ function [u_corr, U_corr_history, q_pred] = ucorr(t, q, sim_data)
put in matrix (Ax <= b) form put in matrix (Ax <= b) form
%} %}
% box constrains % box constraints
% A becomes sort of block-diagonal lb = [];
% A will be at most PREDICTION_HORIZON * 2 * 2 (2: size of T_inv; 2: ub = [];
% accounting for T_inv and -T_inv) by PREDICTION_HORIZON (number of
% vectors in u_corr times the number of elements [2] in each vector)
b_deq = [];
for k=1:pred_hor for k=1:pred_hor
T_inv = T_inv_pred(:,:,k); T_inv = T_inv_pred(:,:,k);
u_track = u_track_pred(:,:,k); u_track = u_track_pred(:,:,k);
d = T_inv*u_track; d = T_inv*u_track;
b_deq = [b_deq; s_ - d; s_ + d]; lb = [lb; -s_-d];
ub = [ub; s_-d];
end end
A_deq = kron(eye(pred_hor), [eye(2); -eye(2)]);
%A_deq
%b_deq
% unknowns
% squared norm of u_corr. H must be identity, % squared norm of u_corr. H must be identity,
% PREDICTION_HORIZON*size(u_corr)
H = eye(pred_hor*2)*2; H = eye(pred_hor*2)*2;
% no linear terms % no linear terms
f = zeros(pred_hor*2, 1); f = zeros(pred_hor*2, 1);
% solve qp problem % solve qp problem
options = optimoptions('quadprog', 'Display', 'off'); options = optimoptions('quadprog', 'Display', 'off');
U_corr = quadprog(H, f, A_deq, b_deq, [],[],[],[],[],options); U_corr = quadprog(H, f, [], [], [],[],lb,ub,[],options);
%U_corr = lsqnonlin(@(pred_hor) ones(pred_hor, 1), U_corr_history(:,:,1), [], [], A_deq, b_deq, [], []);
% reshape the vector of vectors to be an array, each element being % reshape the vector of vectors to be an array, each element being
% u_corr_j as a 2x1 vector % u_corr_j as a 2x1 vector