Disciplined convex programming error: Invalid constraint: {real affine} <= {log-convex} 出错

Could any one hepl me to solve the error?
Disciplined convex programming error:
Invalid constraint: {real affine} <= {log-convex}

出错 <= (第 21 行)
b = newcnstr( evalin( ‘caller’, ‘cvx_problem’, ‘’ ), x, y, ‘<=’ );

出错 trajectory (第 68 行)
talor_j(1,j)<=power(2,dis_j(1,j));

##############
clear;clc
Connect_radius=100;%无人机通信半径
num_visit=size(visited,1);%无人机需要遍历的节点数量
Vmax=30;%速度最大值
Vmin=20;%速度最小值
delta_t=5;%s
last_opt=0;
[row,col]=size(visited);
CI=0.1;
Cu=0.1;
Cj=10^-4;
% for i=2:col-1
% location_now=visted(i,:);
% vist_point=visted(i,:);
visited=[800, 600];
visit_point=[1000 1200];
qc=Sense(visited,visit_point);%在qc时需要考虑waypoint调整
max_round=20;
t=99;
epi=99;
Action_jammer{1,99}=[5 0 0 0];
Action_jammer{2,99}=[0 0 25 0];
Asensor{1,99}=[0 1 0 0];
Asensor{2,99}=[0 0 0 1];
Asensor{3,99}=[0 3 0 0];
Asensor{4,99}=[0 0 0 2];
Asensor{5,99}=[0 0 0 2];
index=1;
waypoint_ini=[1000 1200];
for ite=1:max_round
cvx_begin sdp
variable waypoint(1,2)
variable Uo
variable dis_j(1,jammer_num)
variable dis_i(1,sensor_num)
variable talor_j(1,jammer_num)
variable gju(1,jammer_num)
variable giu(1,sensor_num)
minimize Uo
subject to
% Riuc= Rate(Asensor,Action_jammer,index,t,epi,waypoint);%将初始的waypoint代入,得到干扰机和传感器的功率选择
% taylor_expansion=taylor(Riuc,waypoint,waypoint_ini, ‘Order’, 1);
Fiuc=Rate(Asensor,Action_jammer,index,t,epi,waypoint_ini);
Gju=1inv_pos(Fiuc)inv_pos(0.5rel_entr(2, 1))Jam_Interference(Action_jammer,epi,waypoint_ini);
for j=1:jammer_num
for c= 1:channel_num
dis_j(1,j)= norm(jammer(j,:slight_smile: - waypoint, 2);
dis_j_ini(1,j) = norm(jammer(j,:)-waypoint_ini,2);
GJU(j,c)=Gju(j,c)
(square_pos(dis_j(1,j))-square_pos(dis_j_ini(1,j)));
end
end
sensorsjammer = sens_Interference(Asensor,index,t,waypoint_ini);
giiu=1
inv_pos(Fiuc)inv_pos(0.5rel_entr(2, 1))sensorsjammer;
for i=1:sensor_num
for c= 1:channel_num
sensor_place=cell2mat(sensor(i,1:2));
dis_i(1,i) = norm(sensor_place-waypoint,2);
dis__i_ini(1,i)= pdist2(sensor_place, waypoint_ini, ‘euclidean’);
GIU(i,c)=giiu(i,c)
(square_pos(dis_i(1,i))-square_pos(dis__i_ini(1,i)));
end
end
Riuc_l=Fiuc-sum(sum(GJU))-sum(sum(GIU));
for j=1:jammer_num
talor_j(1,j)=power(2,dis_j_ini(1,j))-2*(waypoint_ini-jammer(j,:))*(waypoint-waypoint_ini)';
talor_j(1,j)<=power(2,dis_j(1,j));
% 对凸表达式进行比较
gju(1,j)<=talor_j(1,j);
end

Welcome to the forum!!

Have you proven this problem is convex? I will assume it is not convex unless you prove otherwise.