Advancing Trajectory Optimization with Approximate Inference: Exploration, Covariance Control and Adaptive Risk
This work addresses the problem of improving trajectory optimization for nonlinear stochastic systems, offering incremental advancements in control as inference methods.
The paper tackled the challenge of discrete-time stochastic optimal control for nonlinear systems under uncertainty by proposing an input inference for control (i2c) algorithm, which demonstrated benefits such as an expert linear Gaussian controller combining open-loop optima and closed-loop variance reduction, inherent adaptive risk sensitivity, and covariance control functionality with minor adjustments.
Discrete-time stochastic optimal control remains a challenging problem for general, nonlinear systems under significant uncertainty, with practical solvers typically relying on the certainty equivalence assumption, replanning and/or extensive regularization. Control as inference is an approach that frames stochastic control as an equivalent inference problem, and has demonstrated desirable qualities over existing methods, namely in exploration and regularization. We look specifically at the input inference for control (i2c) algorithm, and derive three key characteristics that enable advanced trajectory optimization: An `expert' linear Gaussian controller that combines the benefits of open-loop optima and closed-loop variance reduction when optimizing for nonlinear systems, inherent adaptive risk sensitivity from the inference formulation, and covariance control functionality with only a minor algorithmic adjustment.