mforets on gh-pages
build based on 006c239 (compare)
mforets on gh-pages
build based on cffc4ea (compare)
github-actions[bot] on v0.18.3
github-actions[bot] on gh-pages
delete history (compare)
schillic on master
Update Project.toml (compare)
schillic on issubset
schillic on master
extend issubset to unions add _is_intersection_empty for … Merge pull request #630 from Ju… (compare)
mforets on gh-pages
build based on 5f5181a (compare)
schillic on issubset
extend issubset to unions add _is_intersection_empty for … (compare)
mforets on gh-pages
build based on fce866f (compare)
schillic on issubset
add _is_intersection_empty for … (compare)
hi @repolaa,
e == 0
and e == 40
, these will be enabled once you set the value there. so the fact that the solver considers those immediate transitions is correct with the semantics of hybrid automata. you can get around that by introducing a new clock t
that you reset to 0 on each transition and then you require on these transitions in the guards that t
is larger than some positive ε
(say, t > 1e-2
). I guess you need to choose ε > δ
for this to work properly<
) are (I think) treated the same way as non-strict ones. so the condition 0. < e, e < 40.
should always hold and you can just drop itConstrainedLinearMap
for a ConstrainedAffineMap
with zero vectorsorry about the errors you got before. I will look at them in a few weeks, but you caught me at a busy time
e
and p
assume values on a straight line and by writing the enabling conditions for the transitions taking into account that the line is actually a circumference so that multiples of 40 correspond to zero. To do this I am using the mod
operation in the conditions. What I get now is the following error:substitute
to replace the symbolic unwraps.mod
operation symbolically? Here is the updated code https://github.com/repolaa/pursuer-evader/blob/main/PursuitGame.jlPol
are also variables?
mod
?
Pol
to be parametric, can't you just define a function closure?
Pol(e, p, clock) = ...
which instantiates a polyhedron given some values for the variables
By the way if I manage to solve the problem feel free to add it to the JuliaReach docs
That would be awesome; feel free to open issues / open PRs yourself or ask for further help in case of doubts.
LGG09
must be used with full dimensions for (almost all) hybrid systems (as already mentioned before); the error message was not helpful and i opened JuliaReach/ReachabilityAnalysis.jl#612Polyhedra
for that. i will try to add this feature