Jianxiang Feng / 冯健祥

I am a final-year PhD student at Technical University of Munich (TUM) and the Institute of Robotic and Mechtronic (RM), German Aerospace Center (DLR). I am advised by Prof. Rudolph Triebel (DLR&KIT) and affiliated with Munich School of Data Science, where I worked with Prof. Stephan Günnemann. Recently, I start investigating Large/Foundation Models for Robotics as a senior research scientist at Agile Robots AG.

In general, my research interests reside in the intersection of robotics and machine learning with a primary focus on the trustworthy and adaptable learning ability of a robot in an open-world environment. In particular, aiming to equip a robot with introspective capabilities i.e., reliable confidence estimates and an awareness of the internal state of the system scuh as limitation of its knowledge and skills, I am intereseted in levaraging probabilistic Machine Learning methods such as Bayesian Neural Networks, Probabilistic Graphicial Models, flow-based Deep Generative Models, on

     - how to provide such introspection quantitatively: Uncertainty Estimation, Out-Of-Distribution detection, and so on;

     - how to exploit this capability for more trustworthy and cognitive learning-enabled robotic functionalities: perception, assembly planning, active learning, and so on.

I believe that this can lead to more explainable and ultimately safer autonomous robot systems.

P.S. The pronunciation of my first name "Jianxiang" is quite close to "Jensen" (yep, the well known "Jensen inequality" often used in ELBO derivation) but with a different ending :P.

Github     G. Scholar      LinkedIn     Twitter     Instagram

jianxiang.feng at tum dot de

We present a dynamic grasping framework for unknown objects in this work, which uses a five-fingered hand with visual servo control and can compensate for external disturbances..

We evaluated three different ways for quantifying the uncertainty of a VLM failure detector for closed-loop LLM planners.

We propose to equip NFs with efficient but flexible base distributions to overcome the topological constraint for OOD detection in robot learning.

A holistic graphical approach including a graph representation for product assemblies and a policy architecture, Graph Assembly Processing Network, dubbed GRACE to predict assembly sequences in a step-by-step manner.

A density-based method with normalizing flows for feasibility learning in Robotic Assembly based on only feasible examples.

An active learning pipeline to reduce annotation efforts of real data within a Sim-to-Real scenario based on deep Bayesian Neural Networks.

A method for adaptive image classification based on fusing uncertainty estimates from Bayesian Neural Networks as unary potentials within a Conditional Random Field (CRF).

A comprehensive overview of uncertainty estimation in neural networks, reviewing recent advances in the field, highlighting current challenges, and identifying potential research opportunities.

A probabilistic framework to obtain both reliable and fast uncertainty estimates for predictions with Deep Neural Networks (DNNs) based on Sparse Gaussian Processes.

A sparse representation of model uncertainty for Deep Neural Networks (DNNs) where the parameter posterior is approximated with an inverse formulation of the Multivariate Normal Distribution (MND), also known as the information form.

A novel teleoperation system for advancing aerial manipulation in dynamic and unstructured environments based on pose estimation pipelines for the industrial objects of both known and unknown geometries and an active learning pipeline.

Link to the template. Do not scrape the HTML from this page itself, as it includes analytics tags that you do not want on your own website — use the github code instead. Also, consider using Leonid Keselman's Jekyll fork of this page.