Author name cluster

Feng Pan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

IROS Conference 2025 Conference Paper

Ultra-Wideband assisted Visual-Inertial Localization Correction System with Position-Unknown UWB Anchors

Yu Xing
Weixing Li
Feng Pan
Xiaoxue Feng

Given the fact that visual-inertial odometry (VIO) is faced with the challenges of localization drift in the long run, we utilize drift-free Ultra-Wideband (UWB) measurements to eliminate accumulated errors in VIO. Existing UWB-VIO fusion methods are mostly constrained by the accuracy of prior UWB anchor positions. However, in large-scale localization scenarios, the precise locations of UWB anchors are difficult to obtain, and the offline calibration process is complex, significantly limiting flexibility. In this paper, we firstly design a lightweight initialization method based on a dual sliding window structure, which can rapidly obtain initial guesses for the UWB anchor coordinates. After that, we further propose a joint estimation system to refine the anchor coordinates while estimating the correction for VIO. The system combines filter-based and optimization-based methods, which mainly consists of an initialization module and a nonlinear estimator module. The filter in the initialization module provides optimization initial values and covariances, and mutually, the optimization results from the nonlinear estimator provide priors for the filter. Finally, the performance of our proposed approach is verified through both public datasets and real-world experiment. Our project, along with our dataset, has been open-sourced in the form of ROS package and ROS bag.

Details

JBHI Journal 2023 Journal Article

Affinity Feature Strengthening for Accurate, Complete and Robust Vessel Segmentation

Tianyi Shi
Xiaohuan Ding
Wei Zhou
Feng Pan
Zengqiang Yan
Xiang Bai
Xin Yang

Vessel segmentation is crucial in many medical image applications, such as detecting coronary stenoses, retinal vessel diseases and brain aneurysms. However, achieving high pixel-wise accuracy, complete topology structure and robustness to various contrast variations are critical and challenging, and most existing methods focus only on achieving one or two of these aspects. In this paper, we present a novel approach, the affinity feature strengthening network (AFN), which jointly models geometry and refines pixel-wise segmentation features using a contrast-insensitive, multiscale affinity approach. Specifically, we compute a multiscale affinity field for each pixel, capturing its semantic relationships with neighboring pixels in the predicted mask image. This field represents the local geometry of vessel segments of different sizes, allowing us to learn spatial- and scale-aware adaptive weights to strengthen vessel features. We evaluate our AFN on four different types of vascular datasets: X-ray angiography coronary vessel dataset (XCAD), portal vein dataset (PV), digital subtraction angiography cerebrovascular vessel dataset (DSA) and retinal vessel dataset (DRIVE). Extensive experimental results demonstrate that our AFN outperforms the state-of-the-art methods in terms of both higher accuracy and topological metrics, while also being more robust to various contrast changes.

Details DOI

AAAI Conference 2023 Conference Paper

RLEKF: An Optimizer for Deep Potential with Ab Initio Accuracy

Siyu Hu
Wentao Zhang
Qiuchen Sha
Feng Pan
Lin-Wang Wang
Weile Jia
Guangming Tan
Tong Zhao

It is imperative to accelerate the training of neural network force field such as Deep Potential, which usually requires thousands of images based on first-principles calculation and a couple of days to generate an accurate potential energy surface. To this end, we propose a novel optimizer named reorganized layer extended Kalman filtering (RLEKF), an optimized version of global extended Kalman filtering (GEKF) with a strategy of splitting big and gathering small layers to overcome the O(N^2) computational cost of GEKF. This strategy provides an approximation of the dense weights error covariance matrix with a sparse diagonal block matrix for GEKF. We implement both RLEKF and the baseline Adam in our alphaDynamics package and numerical experiments are performed on 13 unbiased datasets. Overall, RLEKF converges faster with slightly better accuracy. For example, a test on a typical system, bulk copper, shows that RLEKF converges faster by both the number of training epochs (x11.67) and wall-clock time (x1.19). Besides, we theoretically prove that the updates of weights converge and thus are against the gradient exploding problem. Experimental results verify that RLEKF is not sensitive to the initialization of weights. The RLEKF sheds light on other AI-for-science applications where training a large neural network (with tons of thousands parameters) is a bottleneck.

PDF Details DOI

AAAI Conference 2007 Conference Paper

Modeling and Learning Vague Event Durations for Temporal Reasoning

Feng Pan

This paper reports on our recent work on modeling and automatically extracting vague, implicit event durations from text (Pan et al. , 2006a, 2006b). It is a kind of commonsense knowledge that can have a substantial impact on temporal reasoning problems. We have also proposed a method of using normal distributions to model judgments that are intervals on a scale and measure their interannotator agreement; this should extend from time to other kinds of vague but substantive information in text and commonsense reasoning.

PDF Details

ICRA Conference 2004 Conference Paper

Robust Procedures for Obtaining Assembly Contact State Extremal Configurations

Feng Pan
Joseph M. Schimmels

Two important components in the selection of an admittance that facilitates force-guided assembly are the identification of: 1) the set of feasible contact states, and 2) the set of configurations that span each contact state, i. e. , the extremal configurations. We present a procedure to automatically generate both sets from CAD models of the assembly parts. In the procedure, all possible combinations of principle contacts are considered when generating hypothesized contact states. The feasibility of each is then evaluated in a genetic algorithm based optimization procedure. The maximum and minimum value of each of the 6 configuration variables spanning each contact state are obtained by again using genetic algorithms. Together, the genetic algorithm approach, the hierarchical data structure containing the states, the relationships among the states, and the extremals within each state are used to provide a reliable means of identifying all feasible contact states and their associated extremal configurations.

Details

ICRA Conference 2003 Conference Paper

Efficient contact state graph generation for assembly applications

Feng Pan
Joseph M. Schimmels

An important aspect in the design of many automated assembly strategies is the ability to automatically generate the set of contact states that may occur during an assembly task. In this paper, we present an efficient means of constructing the set of all geometrically feasible contact states that may occur within a bounded set of misalignments (bounds determined by robot inaccuracy). This set is stored as a graph, referred to as an Assembly Contact State Graph (ACSG), which indicates neighbor relationships between feasible states. An ACSG is constructed without user intervention in two stages. In the first stage, all hypothetical primitive principle contacts (PPCs; all contact states allowing 5 degrees of freedom) are evaluated for geometric feasibility with respect to part-imposed and robot-imposed restrictions on relative positioning (evaluated using optimization). In the second stage, the feasibility of each of the various combinations of PPCs is efficiently evaluated, first using topological existence and uniqueness criteria, then using part-imposed and robot-imposed geometric criteria.

Details