A Julia package for solving constrained trajectory optimization problems:
minimize cost_T(state_T; parameter_T) + sum(cost_t(state_t, action_t; parameter_t))
states, actions
subject to dynamics_t(state_t, action_t, state_t+1; parameter_t), t = 1,...,T-1
constraint_t(state_t, action_t; parameter_t) {<,=} 0, t = 1,...,T
state_lower_t < state_t < state_upper_t, t = 1,...,T
action_lower_t < action_t < action_upper_t, t = 1,...,T-1.
with direct trajectory optimization.
Fast and allocation-free gradients and sparse Jacobians are automatically generated using Symbolics.jl for user-provided costs, constraints, and dynamics. The problem is solved with Ipopt.
Pkg.add("DirectTrajectoryOptimization")