Discrete fully probabilistic design: towards a control pipeline for the synthesis of policies from examples

We present the principled design of a control pipeline for the synthesis of policies from examples data. The pipeline, based on a discretized design, expounds the algorithm introduced in [1] to synthesize policies from examples for constrained, stochastic and nonlinear systems. The pipeline: (i) does not need the constraints to be fulfilled in the possibly noisy example data; (ii) enables control synthesis even when the data are collected from an example system that is different from the one under control. The design is benchmarked on an example that involves controlling an inverted pendulum with actuation constraints. The data that are used to synthesize the policy are collected from a pendulum that: (i) is different from the one under control; (ii) does not satisfy the actuation constraints.