Optimize over Regions 

Solve constrained optimization problems using region constraints. Region constraints are expressed using for a region and can be used alone or together with other constraints. Local and global numerical optimization functions as well as symbolic optimization functions support these new constraints.

Find the minimizing and maximizing point for over a disk.

In[1]:=

X pmin = ArgMin[x^2 - x y + y^2, {x, y} \[Element] Disk[]]

Out[1]=

In[2]:=

X pmax = ArgMax[x^2 - x y + y^2, {x, y} \[Element] Disk[]]

Out[2]=

In[3]:=

X Legended[Show[ContourPlot[x^2 - x y + y^2, {x, y} \[Element] Disk[]], Graphics[{PointSize[Medium], {Red, Point[pmin]}, {Blue, Point[pmax]}}]], SwatchLegend[{Red, Blue}, {"min", "max"}]]

Out[3]=

Find the nearest and farthest points between two regions.

In[4]:=

X \[ScriptCapitalR]1 = Disk[]; \[ScriptCapitalR]2 = Rectangle[{2, 1}, {3, 3}];

In[5]:=

X {p1t, p2t} = NArgMin[EuclideanDistance[p1, p2], {p1 \[Element] \[ScriptCapitalR]1, p2 \[Element] \[ScriptCapitalR]2}]

Out[5]=

In[6]:=

X {q1t, q2t} = NArgMax[EuclideanDistance[p1, p2], {p1 \[Element] \[ScriptCapitalR]1, p2 \[Element] \[ScriptCapitalR]2}]

Out[6]=

In[7]:=

X Graphics[{{LightBlue, \[ScriptCapitalR]1, \[ScriptCapitalR]2}, \ {Dashed, Line[{p1t, p2t}], Red, Point[{p1t, p2t}]}, {Dashed, Line[{q1t, q2t}], Red, Point[{q1t, q2t}]}}]

Out[7]=

Find the nearest and farthest points exactly.

In[8]:=

X \[ScriptCapitalR]1 = Line[{{0, 0, 0}, {1, 1, 1}}]; \[ScriptCapitalR]2 = Ball[{5, 5, 0}, 1];

In[9]:=

X {p1t, p2t} = Simplify@ArgMin[ EuclideanDistance[p1, p2], {p1 \[Element] \[ScriptCapitalR]1, p2 \[Element] \[ScriptCapitalR]2}]

Out[9]=

In[10]:=

X {q1t, q2t} = Simplify@ArgMax[ EuclideanDistance[p1, p2], {p1 \[Element] \[ScriptCapitalR]1, p2 \[Element] \[ScriptCapitalR]2}]

Out[10]=

In[11]:=

X Graphics3D[{{Opacity[ 0.5], \[ScriptCapitalR]1, \[ScriptCapitalR]2}, {Dashed, Line[{p1t, p2t}], Red, Point[{p1t, p2t}]}, {Dashed, Line[{q1t, q2t}], Red, Point[{q1t, q2t}]}}]

Out[11]=