Mobility prediction of off-road Autonomous Ground Vehicles (AGV) in uncertain environments is essential for their model-based mission planning especially in the early design stage. While surrogate-modeling methods have been developed to overcome the computational challenge in simulation-based mobility prediction, it is very challenging for a single surrogate model to accurately capture the complicated vehicle dynamics. With a focus on vertical acceleration of an AGV under off-road conditions, this paper proposes a surrogate modeling approach for AGV mobility prediction using a dynamic ensemble of Nonlinear Autoregressive Network with Exogenous inputs (NARX) models over time. Synthetic vehicle mobility data of an AGV are first collected using a limited number of high-fidelity simulations. The data are then partitioned into different segments using a variational Gaussian mixture model in order to represent different vehicle dynamic behaviors. Based on the partitioned data, multiple surrogate models are constructed under the NARX framework with different numbers of lags. The NARX models are then assembled together dynamically over time to predict the mobility of the AGV under new conditions. A case study demonstrates the advantages of the proposed method over the classical NARX models for AGV mobility prediction.
Optimal navigation of ground vehicles in an off-road setting is a challenging task. One must accurately model the properties of the terrain and reconcile it with vehicle capabilities, while simultaneously addressing mission requirements. An important part of navigation is path planning, the selection of the route a vehicle takes between the start and end points. It is often seen that, given the starting and end points for a vehicle, the optimal path that the vehicle should take varies considerably with the mission requirements. While most commonly used algorithms use a local cost function, mission requirements are typically defined over the entire run of the vehicle. Utility theoretic methods provide a normative tool to model tradeoffs over attributes (mission requirements) that the operator cares about. It is critical therefore, that preferences embedded in the utility function influence the local cost functions used. In this paper, we provide a framework for a feedback-based method to update the parameters of the local cost-function. We do so by using a geodesic-based method for path planning given the terrain inputs, followed by a physics-based simulation of a vehicle to evaluate the attributes. These attributes are then combined into a multiattribute utility function. An optimization-based approach is used to find the parameters of the cost function that maximizes this multiattribute utility. We present our approach on a vehicle navigation example over a terrain acquired from United States Geological Survey data.
Mobility prediction of off-road Autonomous Ground Vehicles (AGV) in uncertain environments is essential for their model-based mission planning especially in the early design stage. While surrogate-modeling methods have been developed to overcome the computational challenge in simulation-based mobility prediction, it is very challenging for a single surrogate model to accurately capture the complicated vehicle dynamics. With a focus on vertical acceleration of an AGV under off-road conditions, this paper proposes a surrogate modeling approach for AGV mobility prediction using a dynamic ensemble of Nonlinear Autoregressive Network with Exogenous inputs (NARX) models over time. Synthetic vehicle mobility data of an AGV are first collected using a limited number of high-fidelity simulations. The data are then partitioned into different segments using a variational Gaussian mixture model in order to represent different vehicle dynamic behaviors. Based on the partitioned data, multiple surrogate models are constructed under the NARX framework with different numbers of lags. The NARX models are then assembled together dynamically over time to predict the mobility of the AGV under new conditions. A case study demonstrates the advantages of the proposed method over the classical NARX models for AGV mobility prediction.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with đź’™ for researchers
Part of the Research Solutions Family.