Author name cluster

Anthony Simeonov

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.

8 papers

1 author row

ICLR Conference 2025 Conference Paper

Diffusion Policy Policy Optimization

Allen Z. Ren
Justin Lidard
Lars Ankile
Anthony Simeonov
Pulkit Agrawal 0001
Anirudha Majumdar
Benjamin Burchfiel
Hongkai Dai

We introduce Diffusion Policy Policy Optimization, DPPO, an algorithmic framework including best practices for fine-tuning diffusion-based policies (e.g. Diffusion Policy) in continuous control and robot learning tasks using the policy gradient (PG) method from reinforcement learning (RL). PG methods are ubiquitous in training RL policies with other policy parameterizations; nevertheless, they had been conjectured to be less efficient for diffusion-based policies. Surprisingly, we show that DPPO achieves the strongest overall performance and efficiency for fine-tuning in common benchmarks compared to other RL methods for diffusion-based policies and also compared to PG fine-tuning of other policy parameterizations. Through experimental investigation, we find that DPPO takes advantage of unique synergies between RL fine-tuning and the diffusion parameterization, leading to structured and on-manifold exploration, stable training, and strong policy robustness. We further demonstrate the strengths of DPPO in a range of realistic settings, including simulated robotic tasks with pixel observations, and via zero-shot deployment of simulation-trained policies on robot hardware in a long-horizon, multi-stage manipulation task.

Details

ICRA Conference 2025 Conference Paper

From Imitation to Refinement - Residual Rl for Precise Assembly

Lars Ankile
Anthony Simeonov
Idan Shenfeld
Marcel Torne
Pulkit Agrawal 0001

Recent advances in Behavior Cloning (BC) have made it easy to teach robots new tasks. However, we find that the ease of teaching comes at the cost of unreliable performance that saturates with increasing data for tasks requiring precision. The performance saturation can be attributed to two critical factors: (a) distribution shift resulting from the use of offline data and (b) the lack of closed-loop corrective control caused by action chucking (predicting a set of future actions executed open-loop) critical for BC performance. Our key insight is that by predicting action chunks, BC policies function more like trajectory “planners” than closedloop controllers necessary for reliable execution. To address these challenges, we devise a simple yet effective method, Resip (Residual for Precise Manipulation), that overcomes the reliability problem while retaining BC's ease of teaching and long-horizon capabilities. Resip augments a frozen, chunked BC model with a fully closed-loop residual policy trained with reinforcement learning (RL) that addresses distribution shifts and introduces closed-loop corrections over open-loop execution of action chunks predicted by the BC trajectory planner. Videos, code, and data: residual-assembly.github.io.

Details

IROS Conference 2024 Conference Paper

JUICER: Data-Efficient Imitation Learning for Robotic Assembly

Lars Ankile
Anthony Simeonov
Idan Shenfeld
Pulkit Agrawal 0001

While learning from demonstrations is powerful for acquiring visuomotor policies, high-performance imitation without large demonstration datasets remains challenging for tasks requiring precise, long-horizon manipulation. This paper proposes a pipeline for improving imitation learning performance with a small human demonstration budget. We apply our approach to assembly tasks that require precisely grasping, reorienting, and inserting multiple parts over long horizons and multiple task phases. Our pipeline combines expressive policy architectures and various techniques for dataset expansion and simulation-based data augmentation. These help expand dataset support and supervise the model with locally corrective actions near bottleneck regions requiring high precision. We demonstrate our pipeline on four furniture assembly tasks in simulation, enabling a manipulator to assemble up to five parts over nearly 2500 time steps directly from RGB images, outperforming imitation and data augmentation baselines. Project website: https://imitation-juicer.github.io/.

Details

ICRA Conference 2024 Conference Paper

Lifelong Robot Learning with Human Assisted Language Planners

Meenal Parakh
Alisha Fong
Anthony Simeonov
Tao Chen 0046
Abhishek Gupta 0004
Pulkit Agrawal 0001

Large Language Models (LLMs) have been shown to act like planners that can decompose high-level instructions into a sequence of executable instructions. However, current LLM-based planners are only able to operate with a fixed set of skills. We overcome this critical limitation and present a method for using LLM-based planners to query new skills and teach robots these skills in a data and time-efficient manner for rigid object manipulation. Our system can re-use newly acquired skills for future tasks, demonstrating the potential of open world and lifelong learning. We evaluate the proposed framework on multiple tasks in simulation and the real world. Videos are available at: https://sites.google.com/mit.edu/halp-robot-learning

Details

ICRA Conference 2023 Conference Paper

Local Neural Descriptor Fields: Locally Conditioned Object Representations for Manipulation

Ethan Chun
Yilun Du
Anthony Simeonov
Tomás Lozano-Pérez
Leslie Pack Kaelbling

A robot operating in a household environment will see a wide range of unique and unfamiliar objects. While a system could train on many of these, it is infeasible to predict all the objects a robot will see. In this paper, we present a method to generalize object manipulation skills acquired from a limited number of demonstrations, to novel objects from unseen shape categories. Our approach, Local Neural Descriptor Fields (L-NDF), utilizes neural descriptors defined on the local geometry of the object to effectively transfer manipulation demonstrations to novel objects at test time. In doing so, we leverage the local geometry shared between objects to produce a more general manipulation framework. We illustrate the efficacy of our approach in manipulating novel objects in novel poses - both in simulation and in the real world. Project website, videos, and code: https://elchun.github.io/lndf/.

Details

ICRA Conference 2022 Conference Paper

Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulation

Anthony Simeonov
Yilun Du
Andrea Tagliasacchi
Joshua B. Tenenbaum
Alberto Rodriguez 0003
Pulkit Agrawal 0001
Vincent Sitzmann

We present Neural Descriptor Fields (NDFs), an object representation that encodes both points and relative poses between an object and a target (such as a robot gripper or a rack used for hanging) via category-level descriptors. We employ this representation for object manipulation, where given a task demonstration, we want to repeat the same task on a new object instance from the same category. We propose to achieve this objective by searching (via optimization) for the pose whose descriptor matches that observed in the demonstration. NDFs are conveniently trained in a self-supervised fashion via a 3D auto-encoding task that does not rely on expert-labeled keypoints. Further, NDFs are SE(3)-equivariant, guaranteeing performance that generalizes across all possible 3D object translations and rotations. We demonstrate learning of manipulation tasks from few (∼5-10) demonstrations both in simulation and on a real robot. Our performance generalizes across both object instances and 6-DoF object poses, and significantly outperforms a recent baseline that relies on 2D descriptors. Project website: https://yilundu.github.io/ndf/

Details

ICRA Conference 2019 Conference Paper

Motion Planning Networks

Ahmed H. Qureshi
Anthony Simeonov
Mayur J. Bency
Michael C. Yip

Fast and efficient motion planning algorithms are crucial for many state-of-the-art robotics applications such as self-driving cars. Existing motion planning methods become ineffective as their computational complexity increases exponentially with the dimensionality of the motion planning problem. To address this issue, we present Motion Planning Networks (MPNet), a neural network-based novel planning algorithm. The proposed method encodes the given workspaces directly from a point cloud measurement and generates the end-to-end collision-free paths for the given start and goal configurations. We evaluate MPNet on various 2D and 3D environments including the planning of a 7 DOF Baxter robot manipulator. The results show that MPNet is not only consistently computationally efficient in all environments but also generalizes to completely unseen environments. The results also show that the computation time of MPNet consistently remains less than 1 second in all presented experiments, which is significantly lower than existing state-of-the-art motion planning algorithms.

Details

ICRA Conference 2018 Conference Paper

Stickman: Towards a Human Scale Acrobatic Robot

Morgan T. Pope
Steven Christensen
David L. Christensen
Anthony Simeonov
Grant Imahara
Günter Niemeyer

Human performers have developed impressive acrobatic techniques over thousands of years of practicing the gymnastic arts. At the same time, robots have started to become more mobile and autonomous, and can begin to imitate these stunts in dramatic and informative ways. We present a simple two degree of freedom robot that uses a gravity-driven pendulum launch and produces a variety of somersaulting stunts. The robot uses an IMU and a laser range-finder to estimate its state mid-flight and actuates to change its motion both on and and off the pendulum. We discuss the dynamics of this behavior in a framework of acrobatic capability and present experimental results.

Details