Sufficiently Accurate Model Learning

Abstract

Data driven models of dynamical systems help planners and controllers to provide more precise and accurate motions. Most model learning algorithms will try to minimize a loss function between the observed data and the model's predictions. This can be improved using prior knowledge about the task at hand, which can be encoded in the form of constraints. This turns the unconstrained model learning problem into a constrained one. These constraints allow models with finite capacity to focus their expressive power on important aspects of the system. This can lead to models that are better suited for certain tasks. This paper introduces the constrained Sufficiently Accurate model learning approach, provides examples of such problems, and presents a theorem on how close some approximate solutions can be. The approximate solution quality will depend on the function parameterization, loss and constraint function smoothness, and the number of samples in model learning.

DRF: A Framework for High-Accuracy Autonomous Driving Vehicle Modeling

“…To overcome the computational issue, Sparse Variational GP (SVGP) method had been developed to incorporate stochastic variations inference to enable mini-batch learning [6] [9] and further, to correct the sensor measurement errors in dynamic modeling [10]. On the other hand, Deep NN were introduced to address dynamic modeling and parameter ID [11] [12]. The successful application of SVGP and NN methods inspired us to integrate them for a more accurate algorithm in this paper.…”