Overload TM functions for Taylor model reachsets

src/Flowpipes/reachsets.jl

@@ -590,9 +590,10 @@ using TaylorModels: TaylorModel1, TaylorN
 """
     TaylorModelReachSet{N} <: AbstractTaylorModelReachSet{N}
 
-Taylor model reach-set represented as a of vector taylor models in one variable
-(the "time" variable) whose coefficients are multivariate polynomials
-(the "space" variables).
+Taylor model reach-set represented as a vector of taylor models in one variable
+(namely, the "time" variable) whose coefficients are multivariate polynomials
+(namely in the "space" variables). It is assumed that the time domain is the same
+for all components.
 
 ### Notes
 
@@ -615,6 +616,22 @@ end
 @inline dim(R::TaylorModelReachSet) = get_numvars()
 @inline vars(R::TaylorModelReachSet) = Tuple(Base.OneTo(length(R.X)),)
 
+# overload getter functions for the taylor model
+# we assume that the first element is representative
+domain(R::TaylorModelReachSet) = domain(first(R.X))
+remainder(R::TaylorModelReachSet) = remainder.(R.X)
+polynomial(R::TaylorModelReachSet) = polynomial.(R.X)
+get_order(R::TaylorModelReachSet) = get_order.(R.X)
+expansion_point(R::TaylorModelReachSet) = [Xi.x0 for Xi in R.X]
+
+# useful constants
+@inline zeroBox(m) = IntervalBox(zeroI, m)
+@inline unitBox(m) = IntervalBox(IA.Interval(0.0, 1.0), m)
+@inline symBox(n::Integer) = IntervalBox(symI, n)
+const zeroI = IA.Interval(0.0) # TODO use number type
+const oneI = IA.Interval(1.0)
+const symI = IA.Interval(-1.0, 1.0)
+
 function shift(R::TaylorModelReachSet, t0::Number)
     return TaylorModelReachSet(set(R), tspan(R) + t0)
 end

src/Initialization/exports.jl

@@ -46,6 +46,12 @@ export
     sup_func, # TODO keep?
     setrep,
     rsetrep,
+    reset_map,
+    guard,
+    source_invariant,
+    target_invariant,
+    # getter functions for Taylor model reach-sets
+    domain, remainder, polynomial, get_order, expansion_point,
     numrsets,
 
 # Concrete operations

src/Initialization/init.jl

@@ -34,6 +34,9 @@ const TimeInterval = IA.Interval{Float64}
 import TaylorModels
 const TM = TaylorModels
 
+# method extensions for Taylor model reach-sets
+import TaylorModels: domain, remainder, polynomial, get_order
+
 @inline function _isapprox(Δt::TimeInterval, Δs::TimeInterval)
     return (inf(Δt) ≈ inf(Δs)) && (sup(Δt) ≈ sup(Δs))
 end

test/algorithms/TMJets.jl

@@ -34,13 +34,21 @@
 end
 
 @testset "TMJets algorithm: linear IVPs" begin
-    prob, tspan = exponential_1d()
-    sol = solve(prob, tspan=tspan, TMJets())
+    prob, dt = exponential_1d()
+    sol = solve(prob, tspan=dt, TMJets())
     @test sol.alg isa TMJets
 
+    # getter functions for a taylor model reach-set
+    R = sol[1]
+    @test domain(R) == tspan(R)
+    @test diam(remainder(R)[1]) < 1e-13
+    @test get_order(R) == [8]
+    @test polynomial(R) isa Vector{Taylor1{TaylorN{Float64}}}
+    @test expansion_point(R) ≈ [IA.Interval(0.0)]
+
     # test intersection with invariant
-    prob, tspan = exponential_1d(invariant=HalfSpace([-1.0], -0.3)) # x >= 0.3
-    sol_inv = solve(prob, tspan=tspan, TMJets())
+    prob, dt = exponential_1d(invariant=HalfSpace([-1.0], -0.3)) # x >= 0.3
+    sol_inv = solve(prob, tspan=dt, TMJets())
     @test [0.3] ∈ overapproximate(sol_inv[end], Zonotope)
     m = length(sol_inv)
     # check that the following reach-set escapes the invariant

test/runtests.jl

@@ -7,6 +7,8 @@ using ReachabilityAnalysis: _isapprox, setrep, rsetrep,
 import IntervalArithmetic
 const IA = IntervalArithmetic
 
+const IA = IntervalArithmetic
+
 # load test models
 include("models/exponential1D.jl")
 include("models/motor.jl")