We propose a reinforcement learning framework
where an agent uses an internal nominal model for stochastic
model predictive control (MPC) while compensating for a
disturbance. Our work builds on the existing risk-aware optimal
control with stochastic differential equations (SDEs) that aims
to deal with such disturbance. However, the risk sensitivity
and the noise strength of the nominal SDE in the riskaware optimal control are often heuristically chosen. In the
proposed framework, the risk-taking policy determines the behavior of the MPC to be risk-seeking (exploration) or riskaverse (exploitation). Specifcally, we employ the risk-aware path integral control that can be implemented as a Monte-Carlo (MC) sampling with fast parallel simulations using a GPU. The MC sampling implementations of the MPC have been successful in robotic applications due to their real-time computation capability. The proposed framework that adapts the noise model and the risk sensitivity outperforms the standard model
predictive path integ
more »
« less
Risk-Aware Model Predictive Control Enabled by Bayesian Learning
The performance of a model predictive controller depends on the accuracy of the objective and prediction model of the system. Although significant efforts have been dedicated to improving the robustness of model predictive control (MPC), they typically do not take a risk-averse perspective. In this paper, we propose a risk-aware MPC framework, which estimates the underlying parameter distribution using online Bayesian learning and derives a risk-aware control policy by reformulating classical MPC problems as Bayesian Risk Optimization (BRO) problems. The consistency of the Bayesian estimator and the convergence of the control policy are rigorously proved. Furthermore, we investigate the consistency requirement and propose a risk monitoring mechanism to guarantee the satisfaction of the consistency requirement. Simulation results demonstrate the effectiveness of the proposed approach.
more »
« less
- NSF-PAR ID:
- 10359107
- Date Published:
- Journal Name:
- Proceedings of 2022 American Control Conference
- Page Range / eLocation ID:
- 108 to 113
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)This paper reports on developing an integrated framework for safety-aware informative motion planning suitable for legged robots. The information-gathering planner takes a dense stochastic map of the environment into account, while safety constraints are enforced via Control Barrier Functions (CBFs). The planner is based on the Incrementally-exploring Information Gathering (IIG) algorithm and allows closed-loop kinodynamic node expansion using a Model Predictive Control (MPC) formalism. Robotic exploration and information gathering problems are inherently path-dependent problems. That is, the information collected along a path depends on the state and observation history. As such, motion planning solely based on a modular cost does not lead to suitable plans for exploration. We propose SAFE-IIG, an integrated informative motion planning algorithm that takes into account: 1) a robot’s perceptual field of view via a submodular information function computed over a stochastic map of the environment, 2) a robot’s dynamics and safety constraints via discrete-time CBFs and MPC for closedloop multi-horizon node expansions, and 3) an automatic stopping criterion via setting an information-theoretic planning horizon. The simulation results show that SAFE-IIG can plan a safe and dynamically feasible path while exploring a dense map.more » « less
-
Training self-driving systems to be robust to the long-tail of driving scenarios is a critical problem. Model-based approaches leverage simulation to emulate a wide range of scenarios without putting users at risk in the real world. One promising path to faithful simulation is to train a forward model of the world to predict the future states of both the environment and the ego-vehicle given past states and a sequence of actions. In this paper, we argue that it is beneficial to model the state of the ego-vehicle, which often has simple, predictable and deterministic behavior, separately from the rest of the environment, which is much more complex and highly multimodal. We propose to model the ego-vehicle using a simple and differentiable kinematic model, while training a stochastic convolutional forward model on raster representations of the state to predict the behavior of the rest of the environment. We explore several configurations of such decoupled models, and evaluate their performance both with Model Predictive Control (MPC) and direct policy learning. We test our methods on the task of highway driving and demonstrate lower crash rates and better stability. The code is available at https://github.com/vladisai/pytorch-PPUU/tree/ICLR2022.more » « less
-
Abstract High-dimensional categorical data are routinely collected in biomedical and social sciences. It is of great importance to build interpretable parsimonious models that perform dimension reduction and uncover meaningful latent structures from such discrete data. Identifiability is a fundamental requirement for valid modeling and inference in such scenarios, yet is challenging to address when there are complex latent structures. In this article, we propose a class of identifiable multilayer (potentially deep) discrete latent structure models for discrete data, termed Bayesian Pyramids. We establish the identifiability of Bayesian Pyramids by developing novel transparent conditions on the pyramid-shaped deep latent directed graph. The proposed identifiability conditions can ensure Bayesian posterior consistency under suitable priors. As an illustration, we consider the two-latent-layer model and propose a Bayesian shrinkage estimation approach. Simulation results for this model corroborate the identifiability and estimatability of model parameters. Applications of the methodology to DNA nucleotide sequence data uncover useful discrete latent features that are highly predictive of sequence types. The proposed framework provides a recipe for interpretable unsupervised learning of discrete data and can be a useful alternative to popular machine learning methods.more » « less
-
The active control of stormwater systems is a potential solution to increased street flooding in low-lying, low-relief coastal cities due to climate change and accompanying sea level rise. Model predictive control (MPC) has been shown to be a successful control strategy generally and as well as for managing urban drainage specifically. This research describes and demonstrates the implementation of MPC for urban drainage systems using open source software (Python and The United States Environmental Protection Agency (EPA) Storm Water Management Model (SWMM5). The system was demonstrated using a simplified use case in which an actively-controlled outlet of a detention pond is simulated. The control of the pond’s outlet influences the flood risk of a downstream node. For each step in the SWMM5 model, a series of policies for controlling the outlet are evaluated. The best policy is then selected using an evolutionary algorithm. The policies are evaluated against an objective function that penalizes primarily flooding and secondarily deviation of the detention pond level from a target level. Freely available Python libraries provide the key functionality for the MPC workflow: step-by-step running of the SWMM5 simulation, evolutionary algorithm implementation, and leveraging parallel computing. For perspective, the MPC results were compared to results from a rule-based approach and a scenario with no active control. The MPC approach produced a control policy that largely eliminated flooding (unlike the scenario with no active control) and maintained the detention pond’s water level closer to a target level (unlike the rule-based approach).more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.23919/ACC53348.2022.9867207
