Danny Drieß

VLDB 2026 Research / reviewers in the wild / expert

dblp:203/4994 · also Danny Driess · DBLP profile ↗

← Back

26ranked-venue papers

10as first author

17since 2021 · last 2025

0000-0002-8258-1659ORCID · corroborated

Domains — the database's venue-derived domains; a paper can count in several

Artificial intelligence and machine learning · 25 · 10 first-author · 16 since 2021Systems, architecture and hardware · 15 · 7 first-author · 7 since 2021Applied, interdisciplinary, general and emerging computing · 1 · 1 since 2021

Year	Publication	Venue	Position
2025	Vision Language Models are In-Context Value Learners abstract Predicting temporal progress from visual trajectories is important for intelligent robots that can learn, adapt, and improve. However, learning such progress estimator, or temporal value function, across different tasks and domains requires both a large amount of diverse data and methods which can scale and generalize. To address these challenges, we present Generative Value Learning (GVL), a universal value function estimator that leverages the world knowledge embedded in vision-language models (VLMs) to predict task progress. Naively asking a VLM to predict values for a video sequence performs poorly due to the strong temporal correlation between successive frames. Instead, GVL poses value estimation as a temporal ordering problem over shuffled video frames; this seemingly more challenging task encourages VLMs to more fully exploit their underlying semantic and temporal grounding capabilities to differentiate frames based on their perceived task progress, consequently producing significantly better value predictions. Without any robot or task specific training, GVL can in-context zero-shot and few-shot predict effective values for more than 300 distinct real-world tasks across diverse robot platforms, including challenging bimanual manipulation tasks. Furthermore, we demonstrate that GVL permits flexible multi-modal in-context learning via examples from heterogeneous tasks and embodiments, such as human videos. The generality of GVL enables various downstream applications pertinent to visuomotor policy learning, including dataset filtering, success detection, and value-weighted regression -- all without any model training or finetuning. Yecheng Jason Ma 0001, Joey Hejna, Chuyuan Fu, Dhruv Shah, Jacky Liang, Sean Kirmani, Peng Xu 0010, Danny Drieß, Ted Xiao, Osbert Bastani, Dinesh Jayaraman, Wenhao Yu 0003, Tingnan Zhang, Dorsa Sadigh, Fei Xia 0002	ICLR	9
2025	Direct Motion Models for Assessing Generated Videos abstract A current limitation of video generative video models is that they generate plausible looking frames, but poor motion — an issue that is not well captured by FVD and other popular methods for evaluating generated videos. Here we go beyond FVD by developing a metric which better measures plausible object interactions and motion. Our novel approach is based on auto-encoding point tracks and yields motion features that can be used to not only compare distributions of videos (as few as one generated and one ground truth, or as many as two datasets), but also for evaluating motion of single videos. We show that using point tracks instead of pixel reconstruction or action recognition features results in a metric which is markedly more sensitive to temporal distortions in synthetic data, and can predict human evaluations of temporal consistency and realism in generated videos obtained from open-source models better than a wide range of alternatives. We also show that by using a point track representation, we can spatiotemporally localize generative video inconsistencies, providing extra interpretability of generated video errors relative to prior work. An overview of the results and link to the code can be found on the project page: trajan-paper.github.io. Kelsey R. Allen, Carl Doersch, Mohammed Suhail, Danny Drieß, Ignacio Rocco, Yulia Rubanova, Thomas Kipf, Mehdi S. M. Sajjadi, Kevin Murphy 0002, João Carreira 0001, Sjoerd van Steenkiste	ICML	5
2025	Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Models abstract Generalist robots that can perform a range of different tasks in open-world settings must be able to not only reason about the steps needed to accomplish their goals, but also process complex instructions, prompts, and even feedback during task execution. Intricate instructions (e.g., "Could you make me a vegetarian sandwich?" or "I don’t like that one") require not just the ability to physically perform the individual steps, but the ability to situate complex commands and feedback in the physical world. In this work, we describe a system that uses vision-language models in a hierarchical structure, first reasoning over complex prompts and user feedback to deduce the most appropriate next step to fulfill the task, and then performing that step with low-level actions. In contrast to direct instruction following methods that can fulfill simple commands ("pick up the cup"), our system can reason through complex prompts and incorporate situated feedback during task execution ("that’s not trash"). We evaluate our system across three robotic platforms, including single-arm, dual-arm, and dual-arm mobile robots, demonstrating its ability to handle tasks such as cleaning messy tables, making sandwiches, and grocery shopping. Videos are available at https://www.pi.website/research/hirobot Lucy Xiaoyang Shi, Brian Ichter, Michael Equi, Liyiming Ke, Karl Pertsch, James Tanner, Anna Walling, Haohuan Wang, Niccolo Fusai, Adrian Li-Bell, Danny Drieß, Lachy Groom, Sergey Levine, Chelsea Finn	ICML	12
2025	RT-Affordance: Affordances are Versatile Intermediate Representations for Robot Manipulation abstract We explore how intermediate policy representations can facilitate generalization by providing guidance on how to perform manipulation tasks. Existing representations such as language, goal images, and trajectory sketches have been shown to be helpful, but these representations either do not provide enough context or provide over-specified context that yields less robust policies. We propose conditioning policies on affordances, which capture the pose of the robot at key stages of the task. Affordances offer expressive yet lightweight abstractions, are easy for users to specify, and facilitate efficient learning by transferring knowledge from large internet datasets. Our method, RT-Affordance, is a hierarchical model that first proposes an affordance plan given the task language, and then conditions the policy on this affordance plan to perform manipulation. Our model can flexibly bridge heterogeneous sources of supervision including large web datasets and robot trajectories. We additionally train our model on cheap-to-collect in-domain affordance images, allowing us to learn new tasks without collecting any additional costly robot trajectories. We show on a diverse set of novel tasks how RT-Affordance exceeds the performance of existing methods by over 50 %, and we empirically demonstrate that affordances are robust to novel settings. Videos available at https://snasiriany. me/rt-affordance Soroush Nasiriany, Sean Kirmani, Tianli Ding, Laura Smith 0001, Yuke Zhu, Danny Drieß, Dorsa Sadigh, Ted Xiao	ICRA	6
2025	Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better abstract Vision-language-action (VLA) models provide a powerful approach to training control policies for physical systems, such as robots, by combining end-to-end learning with transfer of semantic knowledge from web-scale vision-language model (VLM) training. However, the constraints of real-time control are often at odds with the design of VLMs: the most powerful VLMs have tens or hundreds of billions of parameters, presenting an obstacle to real-time inference, and operate on discrete tokens rather than the continuous-valued outputs that are required for controlling robots. To address this challenge, recent VLA models have used specialized modules for efficient continuous control, such as action experts or continuous output heads, which typically require adding new untrained parameters to the pretrained VLM backbone. While these modules improve real-time and control capabilities, it remains an open question whether they preserve or degrade the semantic knowledge contained in the pretrained VLM, and what effect they have on the VLA training dynamics. In this paper, we study this question in the context of VLAs that include a continuous diffusion or flow matching action expert, showing that naively including such experts significantly harms both training speed and knowledge transfer. We provide an extensive analysis of various design choices, their impact on performance and knowledge transfer, and propose a technique for insulating the VLM backbone during VLA training that mitigates this issue. Videos are available at https://pi.website/research/knowledge_insulation and open-source model weights are available at https://github.com/Physical-Intelligence/openpi. Danny Drieß, Jost Tobias Springenberg, Brian Ichter, Lili Yu, Adrian Li-Bell, Karl Pertsch, Allen Z. Ren, Homer Walke, Lucy Xiaoyang Shi, Sergey Levine	NeurIPS	1
2024	PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs abstract Vision language models (VLMs) have shown impressive capabilities across a variety of tasks, from logical reasoning to visual understanding. This opens the door to richer interaction with the world, for example robotic control. However, VLMs produce only textual outputs, while robotic control and other spatial tasks require outputting continuous coordinates, actions, or trajectories. How can we enable VLMs to handle such settings without fine-tuning on task-specific data? In this paper, we propose a novel visual prompting approach for VLMs that we call Prompting with Iterative Visual Optimization (PIVOT), which casts tasks as iterative visual question answering. In each iteration, the image is annotated with a visual representation of proposals that the VLM can refer to (e.g., candidate robot actions, localizations, or trajectories). The VLM then selects the best ones for the task. These proposals are iteratively refined, allowing the VLM to eventually zero in on the best available answer. We investigate PIVOT on real-world robotic navigation, real-world manipulation from images, instruction following in simulation, and additional spatial inference tasks such as localization. We find, perhaps surprisingly, that our approach enables zero-shot control of robotic systems without any robot training data, navigation in a variety of environments, and other capabilities. Although current performance is far from perfect, our work highlights potentials and limitations of this new regime and shows a promising approach for Internet-Scale VLMs in robotic and spatial reasoning domains. Soroush Nasiriany, Fei Xia 0002, Wenhao Yu 0003, Ted Xiao, Jacky Liang, Ishita Dasgupta 0001, Annie Xie, Danny Drieß, Ayzaan Wahid, Tingnan Zhang, Tsang-Wei Edward Lee, Kuang-Huei Lee, Peng Xu 0010, Sean Kirmani, Yuke Zhu, Andy Zeng 0001, Karol Hausman, Nicolas Heess, Chelsea Finn, Sergey Levine, Brian Ichter	ICML	8
2024	Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration abstract Large, high-capacity models trained on diverse datasets have shown remarkable successes on efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led to a consolidation of pretrained models, with general pretrained backbones serving as a starting point for many applications. Can such a consolidation happen in robotics? Conventionally, robotic learning methods train a separate model for every application, every robot, and even every environment. Can we instead train "generalist" X-robot policy that can be adapted efficiently to new robots, tasks, and environments? In this paper, we provide datasets in standardized data formats and models to make it possible to explore this possibility in the context of robotic manipulation, alongside experimental results that provide an example of effective X-robot policies. We assemble a dataset from 22 different robots collected through a collaboration between 21 institutions, demonstrating 527 skills (160266 tasks). We show that a high-capacity model trained on this data, which we call RT-X, exhibits positive transfer and improves the capabilities of multiple robots by leveraging experience from other platforms. The project website is robotics-transformer-x.github.io. Abigail O'Neill, Abhiram Maddukuri, Abhishek Gupta 0004, Abhishek Padalkar, Abraham Lee, Acorn Pooley, Agrim Gupta, Ajay Mandlekar, Ajinkya Jain, Albert Tung, Alex Bewley, Alex Irpan, Alexander Khazatsky, Anant Rai, Anchit Gupta, Andrew E. Wang, Anikait Singh, Animesh Garg, Aniruddha Kembhavi, Annie Xie, Anthony Brohan, Antonin Raffin, Archit Sharma, Arefeh Yavary, Arhan Jain, Ashwin Balakrishna, Ayzaan Wahid, Ben Burgess-Limerick, Bernhard Schölkopf, Blake Wulfe, Brian Ichter, Cewu Lu, Charles Xu 0003, Charlotte Le, Chelsea Finn, Chen Wang 0053, Chenfeng Xu, Cheng Chi 0001, Chenguang Huang, Christine Chan, Christopher Agia, Chuer Pan, Chuyuan Fu, Coline Devin, Danfei Xu, Daniel Morton, Danny Drieß, Daphne Chen, Deepak Pathak, Dhruv Shah, Dieter Büchler, Dinesh Jayaraman, Dmitry Kalashnikov, Dorsa Sadigh, Edward Johns, Ethan Paul Foster, Fangchen Liu, Federico Ceola, Fei Xia 0002, Feiyu Zhao, Freek Stulp, Gaoyue Zhou, Gaurav S. Sukhatme, Gautam Salhotra, Gilbert Feng, Giulio Schiavi, Glen Berseth, Gregory Kahn, Guanzhi Wang, Hao Su 0001, Haoshu Fang, Henghui Bao, Heni Ben Amor, Henrik I. Christensen, Hiroki Furuta, Homer Walke, Hongjie Fang, Huy Ha, Igor Mordatch, Ilija Radosavovic, Isabel Leal, Jacky Liang, Jad Abou-Chakra, Jaehyung Kim 0001, Jaimyn Drake, Jan Peters 0001, Jan Schneider 0007, Jasmine Hsu, Jeannette Bohg, Jeffrey T. Bingham, Jensen Gao, Jiaheng Hu, Jiajun Wu 0001, Jiankai Sun, Jianlan Luo, Jiayuan Gu, Jie Tan 0001, Jihoon Oh, Jimmy Wu, Jingpei Lu, Jitendra Malik, João Silvério, Joey Hejna, Jonathan Booher, Jonathan Tompson, Jonathan Yang, Jordi Salvador, Joseph J. Lim, Junhyek Han, Kanishka Rao, Karl Pertsch, Karol Hausman, Keegan Go, Keerthana Gopalakrishnan, Kenneth Y. Goldberg, Kendra Byrne, Kenneth Oslund, Kento Kawaharazuka, Kevin Black, Kevin Zhang 0002, Kiana Ehsani, Kiran Lekkala, Kirsty Ellis, Krishan Rana, Krishnan Srinivasan, Kuan Fang, Kunal Pratap Singh, Kuo-Hao Zeng, Kyle Hatch, Kyle Hsu, Laurent Itti, Yunliang Chen 0001, Lerrel Pinto, Li Fei-Fei 0001, Liam Tan, Linxi Fan, Lionel Ott, Lisa Lee, Luca Weihs, Magnum Chen, Marion Lepert, Marius Memmel, Masayoshi Tomizuka, Masha Itkina, Mateo Guaman Castro, Max Spero, Maximilian Du, Michael Ahn, Michael C. Yip, Mingtong Zhang 0003, Mingyu Ding, Minho Heo, Mohan Kumar Srirama, Mohit Sharma 0001, Moo Jin Kim, Naoaki Kanazawa, Nicklas Hansen 0001, Nicolas Heess, Nikhil J. Joshi, Niko Sünderhauf, Norman Di Palo, Nur Muhammad Shafiullah, Oier Mees, Oliver Kroemer, Osbert Bastani, Pannag R. Sanketi, Patrick Tree Miller, Patrick Yin, Paul Wohlhart, Peng Xu 0010, Peter David Fagan, Peter Mitrano, Pierre Sermanet, Pieter Abbeel, Priya Sundaresan, Qiuyu Chen, Rafael Rafailov, Ria Doshi, Roberto Martin Martin, Rohan Baijal, Rosario Scalise, Rose Hendrix, Roy Lin, Runjia Qian, Russell Mendonca, Rutav Shah, Ryan Hoque, Ryan Julian, Samuel Bustamante-Gomez, Sean Kirmani, Sergey Levine, Sherry Moore, Shikhar Bahl, Shivin Dass, Shubham D. Sonawani, Shuran Song, Sichun Xu, Siddhant Haldar, Siddharth Karamcheti, Simeon Adebola, Simon Guist, Soroush Nasiriany, Stefan Schaal, Stefan Welker, Stephen Tian, Subramanian Ramamoorthy, Sudeep Dasari, Suneel Belkhale, Sungjae Park, Suraj Nair 0003, Suvir Mirchandani, Takayuki Osa, Tanmay Gupta, Tatsuya Harada, Tatsuya Matsushima, Ted Xiao, Thomas Kollar, Tianhe Yu, Tianli Ding, Todor Davchev, Tony Z. Zhao, Travis Armstrong, Trevor Darrell, Trinity Chung, Vidhi Jain, Vincent Vanhoucke, Wolfram Burgard, Xiaolong Wang 0004, Xinghao Zhu, Xinyang Geng, Liangwei Xu, Yecheng Jason Ma 0001, Yejin Kim 0003, Yevgen Chebotar, Yilin Wu 0003, Yonatan Bisk, Yoonyoung Cho, Youngwoon Lee, Yuchen Cui, Yueh-Hua Wu, Yujin Tang, Yuke Zhu, Yunchu Zhang, Yunfan Jiang 0001, Yunshuang Li, Yunzhu Li, Yusuke Iwasawa, Yutaka Matsuo, Zehan Ma, Zichen Jeff Cui, Zichen Zhang 0016, Zipeng Lin	ICRA	50
2023	PaLM-E: An Embodied Multimodal Language Model abstract Large language models excel at a wide range of complex tasks. However, enabling general inference in the real world, e.g. for robotics problems, raises the challenge of grounding. We propose embodied language models to directly incorporate real-world continuous sensor modalities into language models and thereby establish the link between words and percepts. Input to our embodied language model are multimodal sentences that interleave visual, continuous state estimation, and textual input encodings. We train these encodings end-to-end, in conjunction with a pre-trained large language model, for multiple embodied tasks including sequential robotic manipulation planning, visual question answering, and captioning. Our evaluations show that PaLM-E, a single large embodied multimodal model, can address a variety of embodied reasoning tasks, from a variety of observation modalities, on multiple embodiments, and further, exhibits positive transfer: the model benefits from diverse joint training across internet-scale language, vision, and visual-language domains. Our largest model with 562B parameters, in addition to being trained on robotics tasks, is a visual-language generalist with state-of-the-art performance on OK-VQA, and retains generalist language capabilities with increasing scale. Danny Drieß, Fei Xia 0002, Mehdi S. M. Sajjadi, Corey Lynch, Aakanksha Chowdhery, Brian Ichter, Ayzaan Wahid, Jonathan Tompson, Tianhe Yu, Wenlong Huang, Yevgen Chebotar, Pierre Sermanet, Daniel Duckworth, Sergey Levine, Vincent Vanhoucke, Karol Hausman, Marc Toussaint, Klaus Greff, Andy Zeng 0001, Igor Mordatch, Peter R. Florence	ICML	1
2023	Learning Feasibility of Factored Nonlinear Programs in Robotic Manipulation Planning abstract A factored Nonlinear Program (Factored-NLP) explicitly models the dependencies between a set of continuous variables and nonlinear constraints, providing an expressive formulation for relevant robotics problems such as manipulation planning or simultaneous localization and mapping. When the problem is over-constrained or infeasible, a fundamental issue is to detect a minimal subset of variables and constraints that are infeasible. Previous approaches require solving several nonlinear programs, incrementally adding and removing constraints, and are thus computationally expensive. In this paper, we propose a graph neural architecture that predicts which variables and constraints are jointly infeasible. The model is trained with a dataset of labeled subgraphs of Factored-NLPs, and importantly, can make useful predictions on larger factored nonlinear programs than the ones seen during training. We evaluate our approach in robotic manipulation planning, where our model is able to generalize to longer manipulation sequences involving more objects and robots, and different geometric environments. The experiments show that the learned model accelerates general algorithms for conflict extraction (by a factor of 50) and heuristic algorithms that exploit expert knowledge (by a factor of 4). Joaquim Ortiz de Haro, Jung-Su Ha, Danny Drieß, Erez Karpas, Marc Toussaint	ICRA	3
2023	Grounded Decoding: Guiding Text Generation with Grounded Models for Embodied Agents abstract Recent progress in large language models (LLMs) has demonstrated the ability to learn and leverage Internet-scale knowledge through pre-training with autoregressive models. Unfortunately, applying such models to settings with embodied agents, such as robots, is challenging due to their lack of experience with the physical world, inability to parse non-language observations, and ignorance of rewards or safety constraints that robots may require. On the other hand, language-conditioned robotic policies that learn from interaction data can provide the necessary grounding that allows the agent to be correctly situated in the real world, but such policies are limited by the lack of high-level semantic understanding due to the limited breadth of the interaction data available for training them. Thus, if we want to make use of the semantic knowledge in a language model while still situating it in an embodied setting, we must construct an action sequence that is both likely according to the language model and also realizable according to grounded models of the environment. We frame this as a problem similar to probabilistic filtering: decode a sequence that both has high probability under the language model and high probability under a set of grounded model objectives. We demonstrate how such grounded models can be obtained across three simulation and real-world domains, and that the proposed decoding strategy is able to solve complex, long-horizon embodiment tasks in a robotic setting by leveraging the knowledge of both models. Wenlong Huang, Fei Xia 0002, Dhruv Shah, Danny Drieß, Andy Zeng 0001, Yao Lu 0006, Peter R. Florence, Igor Mordatch, Sergey Levine, Karol Hausman, Brian Ichter	NeurIPS	4
2023	Long-Horizon Multi-Robot Rearrangement Planning for Construction Assembly abstract Robotic construction assembly planning aims to find feasible assembly sequences as well as the corresponding robot-paths and can be seen as a special case of task and motion planning (TAMP). As construction assembly can well be parallelized, it is desirable to plan for multiple robots acting concurrently. Solving TAMP instances with many robots and over a long time-horizon is challenging due to coordination constraints, and the difficulty of choosing the right task assignment. We present a planning system which enables parallelization of complex task and motion planning problems by iteratively solving smaller subproblems. Combining optimization methods to jointly solve for manipulation constraints with a sampling-based bi-directional space-time path planner enables us to plan cooperative multi-robot manipulation with unknown arrival-times. Thus, our solver allows for completing subproblems and tasks with differing timescales and synchronizes them effectively. We demonstrate the approach on multiple construction case-studies to show the robustness over long planning horizons and scalability to many objects and agents. Finally, we also demonstrate the execution of the computed plans on two robot arms to showcase the feasibility in the real world. Valentin N. Hartmann, Andreas Orthey, Danny Drieß, Ozgur S. Oguz, Marc Toussaint	IEEE Trans. Robotics	3
2022	FC3: Feasibility-Based Control Chain Coordination abstract Hierarchical coordination of controllers often uses symbolic state representations that fully abstract their underlying low-level controllers, treating them as “black boxes” to the symbolic action abstraction. This paper proposes a framework to realize robust behavior, which we call Feasibility-based Control Chain Coordination (FC3). Our controllers expose the geometric features and constraints they operate on. Based on this, FC3can reason over the controllers' feasibility and their sequence feasibility. For a given task, FC3first automatically constructs a library of potential controller chains using a symbolic action tree, which is then used to coordinate controllers in a chain, evaluate task feasibility, as well as switching between controller chains if necessary. In several real-world experiments we demonstrate FC3, s robustness and awareness of the task's feasibility through its own actions and gradual responses to different interferences. Jason Harris, Danny Drieß, Marc Toussaint	IROS	2
2022	Sequence-of-Constraints MPC: Reactive Timing-Optimal Control of Sequential Manipulation abstract Task and Motion Planning has made great progress in solving hard sequential manipulation problems. However, a gap between such planning formulations and control methods for reactive execution remains. In this paper we pro-pose a model predictive control approach dedicated to robustly execute a single sequence of constraints, which corresponds to a discrete decision sequence of a TAMP plan. We decompose the overall control problem into three sub-problems (solving for sequential waypoints, their timing, and a short receding horizon path) that each is a non-linear program solved online in each MPC cycle. The resulting control strategy can account for long-term interdependencies of constraints and reactively plan for a timing-optimal transition through all constraints. We additionally propose phase backtracking when running constraints of the current phase cannot be fulfilled, leading to a fluent re-initiation behavior that is robust to perturbations and interferences by an experimenter. Marc Toussaint, Jason Harris, Jung-Su Ha, Danny Drieß, Wolfgang Hönig	IROS	4
2022	Reinforcement Learning with Neural Radiance Fields abstract It is a long-standing problem to find effective representations for training reinforcement learning (RL) agents. This paper demonstrates that learning state representations with supervision from Neural Radiance Fields (NeRFs) can improve the performance of RL compared to other learned representations or even low-dimensional, hand-engineered state information. Specifically, we propose to train an encoder that maps multiple image observations to a latent space describing the objects in the scene. The decoder built from a latent-conditioned NeRF serves as the supervision signal to learn the latent space. An RL algorithm then operates on the learned latent space as its state representation. We call this NeRF-RL. Our experiments indicate that NeRF as supervision leads to a latent space better suited for the downstream RL tasks involving robotic object manipulations like hanging mugs on hooks, pushing objects, or opening doors.Video: https://dannydriess.github.io/nerf-rl Danny Drieß, Ingmar Schubert, Peter R. Florence, Yunzhu Li, Marc Toussaint	NeurIPS	1
2021	Learning Geometric Reasoning and Control for Long-Horizon Tasks from Visual Input abstract Long-horizon manipulation tasks require joint reasoning over a sequence of discrete actions and their associated continuous control parameters. While Task and Motion Planning (TAMP) approaches are capable of generating motion plans that account for this joint reasoning, they usually assume full knowledge about the environment (e.g. in terms of shapes, poses of objects) and often require computation times not suitable for real-time control.To overcome this, we propose a learning framework where a high-level reasoning network predicts, based on an image of the scene, a sequence of discrete actions and the parameter values of their associated low-level controllers. These controllers are parameterized in terms of a learned energy function, leading to time-invariant controllers for each phase. We train the whole framework end-to-end using a dataset of TAMP solutions computed using Logic Geometric Programming. A key feature is that the reasoning network determines the parameters of the controllers jointly, such that the overall task can be solved. Despite having no explicit representation of the geometry nor pose of the objects in the scene, our network is still able to accomplish geometrically precise manipulation tasks, including handovers and an accurate pointing task where the parameters of early actions are tightly coupled with those of later actions. Video: https://youtu.be/AcPWRTkr3_g Danny Drieß, Jung-Su Ha, Russ Tedrake, Marc Toussaint	ICRA	1
2021	Deep 6-DoF Tracking of Unknown Objects for Reactive Grasping abstract Robotic manipulation of unknown objects is an important field of research. Practical applications occur in many real-world settings where robots need to interact with an unknown environment. We tackle the problem of reactive grasping by proposing a method for unknown object tracking, grasp point sampling and dynamic trajectory planning. Our object tracking method combines Siamese Networks with an Iterative Closest Point approach for pointcloud registration into a method for 6-DoF unknown object tracking. The method does not require further training and is robust to noise and occlusion. We propose a robotic manipulation system, which is able to grasp a wide variety of formerly unseen objects and is robust against object perturbations and inferior grasping points. Marc Tuscher, Julian Hörz, Danny Drieß, Marc Toussaint	ICRA	3
2021	Learning to Execute: Efficient Learning of Universal Plan-Conditioned Policies in Robotics abstract Applications of Reinforcement Learning (RL) in robotics are often limited by high data demand. On the other hand, approximate models are readily available in many robotics scenarios, making model-based approaches like planning a data-efficient alternative. Still, the performance of these methods suffers if the model is imprecise or wrong. In this sense, the respective strengths and weaknesses of RL and model-based planners are complementary. In the present work, we investigate how both approaches can be integrated into one framework that combines their strengths. We introduce Learning to Execute (L2E), which leverages information contained in approximate plans to learn universal policies that are conditioned on plans. In our robotic manipulation experiments, L2E exhibits increased performance when compared to pure RL, pure planning, or baseline methods combining learning and planning. Ingmar Schubert, Danny Drieß, Ozgur S. Oguz, Marc Toussaint	NeurIPS	2
2020	Adaptive Discretization for Evaluation of Probabilistic Cost Functions abstract In many real-world planning applications, e.g. dynamic design of experiments, autonomous driving and robot manipulation, it is necessary to evaluate candidate movement paths with respect to a safety cost function. Here, the continuous candidate paths need to be discretized first and, subsequently, evaluated onthe discretization points. The resulting quality of planned paths, thus, highly depends on the definition of the safety cost functions, and the resolution of the discretization. In this paper, we propose an approach for evaluating continuous candidate paths by employing an adaptive discretization scheme, with a probabilistic cost function learned from observations. The obtained path is then guaranteed to be epsilon-safe, i.e. the remaining risk of still finding an unsafe point on the trajectory is smaller than epsilon. The proposed approach is investigated theoretically, as well as empirically validated on several robotic path planning scenarios. Christoph Zimmer, Danny Drieß, Mona Meister, Duy Nguyen-Tuong	AISTATS	2
2020	Deep Visual Heuristics: Learning Feasibility of Mixed-Integer Programs for Manipulation Planning abstract In this paper, we propose a deep neural network that predicts the feasibility of a mixed-integer program from visual input for robot manipulation planning. Integrating learning into task and motion planning is challenging, since it is unclear how the scene and goals can be encoded as input to the learning algorithm in a way that enables to generalize over a variety of tasks in environments with changing numbers of objects and goals. To achieve this, we propose to encode the scene and the target object directly in the image space.Our experiments show that our proposed network generalizes to scenes with multiple objects, although during training only two objects are present at the same time. By using the learned network as a heuristic to guide the search over the discrete variables of the mixed-integer program, the number of optimization problems that have to be solved to find a feasible solution or to detect infeasibility can greatly be reduced. Danny Drieß, Ozgur S. Oguz, Jung-Su Ha, Marc Toussaint	ICRA	1
2020	A Probabilistic Framework for Constrained Manipulations and Task and Motion Planning under Uncertainty abstract Logic-Geometric Programming (LGP) is a powerful motion and manipulation planning framework, which represents hierarchical structure using logic rules that describe discrete aspects of problems, e.g., touch, grasp, hit, or push, and solves the resulting smooth trajectory optimization. The expressive power of logic allows LGP for handling complex, large-scale sequential manipulation and tool-use planning problems. In this paper, we extend the LGP formulation to stochastic domains. Based on the control-inference duality, we interpret LGP in a stochastic domain as fitting a mixture of Gaussians to the posterior path distribution, where each logic pro le defines a single Gaussian path distribution. The proposed framework enables a robot to prioritize various interaction modes and to acquire interesting behaviors such as contact exploitation for uncertainty reduction, eventually providing a composite control scheme that is reactive to disturbance. Jung-Su Ha, Danny Drieß, Marc Toussaint	ICRA	2
2020	Robust Task and Motion Planning for Long-Horizon Architectural Construction Planning abstract Integrating robotic systems in architectural and construction processes is of core interest to increase the efficiency of the building industry. Automated planning for such systems enables design analysis tools and facilitates faster design iteration cycles for designers and engineers. However, generic task-and-motion planning (TAMP) for long-horizon construction processes is beyond the capabilities of current approaches. In this paper, we develop a multi-agent TAMP framework for long horizon problems such as constructing a full-scale building. To this end we extend the Logic-Geometric Programming framework by sampling-based motion planning, a limited horizon approach, and a task-specific structural stability optimization that allow an effective decomposition of the task. We show that our framework is capable of constructing a large pavilion built from several hundred geometrically unique building elements from start to end autonomously. Valentin N. Hartmann, Ozgur S. Oguz, Danny Drieß, Marc Toussaint, Achim Menges	IROS	3
2019	Active Multi-Contact Continuous Tactile Exploration with Gaussian Process Differential Entropy abstract In the present work, we propose an active tactile exploration framework to obtain a surface model of an unknown object utilizing multiple contacts simultaneously. To incorporate these multiple contacts, the exploration strategy is based on the differential entropy of the underlying Gaussian process implicit surface model, which formalizes the exploration with multiple contacts within an information theoretic context and additionally allows for nonmyopic multi-step planning. In contrast to many previous approaches, the robot continuously slides along the surface with its end-effectors to gather the tactile stimuli, instead of touching it at discrete locations. This is realized by closely integrating the surface model into the compliant controller framework. Furthermore, we extend our recently proposed sliding based tactile exploration approach to handle non-convex objects. In the experiments, it is shown that multiple contacts simultaneously leads to a more efficient exploration of complex, non-convex objects, not only in terms of time, but also with respect to the total moved distance of all end-effectors. Finally, we demonstrate our methodology with a real PR2 robot that explores an object with both of its arms. Danny Drieß, Daniel Hennes, Marc Toussaint	ICRA	1
2019	Active Inverse Model Learning with Error and Reachable Set Estimates abstract In this work, we propose a framework to learn an inverse model of redundant systems. We address three problems. By formalizing what it actually means to learn an inverse model, we derive a method where the inverse model, represented as a neural network, is learned by minimizing an upper bound on the real performance error, which is provided by a forward model (kernel regression or Gaussian process) learned on the currently available data. Most machine learning methods focus on learning the mapping of the function. For inverse models, it is, however, crucial to know the reachable set of the true forward model, since this becomes the domain of the inverse. Therefore, we secondly propose a method to estimate the reachable set of the system. Finally, we develop an active exploration strategy that is based on maximizing a lower bound on the true fill-distance to efficiently generate the data in the high dimensional input space. A key feature of our method is that the resulting learned inverse model provides error bounds on its performance.From an application point of view, this work is motivated by learning to control musculoskeletal systems. In the experiments, we show for both a simulated model of a human arm with six muscles and a real muscle-driven robot that the proposed method is able to learn the reachable set of these systems as well as a policy that enables to accurately control the position. Danny Drieß, Syn Schmitt, Marc Toussaint	IROS	1
2018	Learning to Control Redundant Musculoskeletal Systems with Neural Networks and SQP: Exploiting Muscle Properties abstract Modeling biomechanical musculoskeletal systems reveals that the mapping from muscle stimulations to movement dynamics is highly nonlinear and complex, which makes it difficult to control those systems with classical techniques. In this work, we not only investigate whether machine learning approaches are capable of learning a controller for such systems. We are especially interested in the question if the structure of the musculoskeletal apparatus exhibits properties that are favorable for the learning task. In particular, we consider learning a control policy from target positions to muscle stimulations. To account for the high actuator redundancy of biomechanical systems, our approach uses a learned forward model represented by a neural network and sequential quadratic programming to obtain the control policy, which also enables us to alternate the co-contraction level and hence allows to change the stiffness of the system and to include optimality criteria like small muscle stimulations. Experiments on both a simulated musculoskeletal model of a human arm and a real biomimetic muscle-driven robot show that our approach is able to learn an accurate controller despite high redundancy and nonlinearity, while retaining sample efficiency. Danny Drieß, Heiko Zimmermann, Simon Wolfen, Dan Suissa, Daniel F. B. Haeufle, Daniel Hennes, Marc Toussaint, Syn Schmitt	ICRA	1
2017	Constrained Bayesian optimization of combined interaction force/task space controllers for manipulations abstract In this paper, we address the problem of how a robot can optimize parameters of combined interaction force/task space controllers under a success constraint in an active way. To enable the robot to explore its environment robustly, safely and without the risk of damaging anything, suitable control concepts have to be developed that enable compliant and force control in situations that are afflicted with high uncertainties. Instances of such concepts are impedance, operational space or hybrid control. However, the parameters of these controllers have to be tuned precisely in order to achieve reasonable performance, which is inherently challenging, as often no sufficient model of the environment is available. To overcome this, we propose to use constrained Bayesian optimization to enable the robot to tune its controller parameters autonomously. Unlike other controller tuning methods, this method allows us to include a success constraint into the optimization. Further, we introduce novel performance measures for compliant, force controlled robots. In real world experiments we show that our approach is able to optimize the parameters for a task that consists of establishing and maintaining contact between the robot and the environment efficiently and successfully. Danny Drieß, Peter Englert, Marc Toussaint	ICRA	1
2017	Active learning with query paths for tactile object shape exploration abstract In the present work, we propose an active learning framework based on optimal query paths to efficiently address the problem of tactile object shape exploration. Most previous approaches perform active touch probing at discrete query points, which leads to inefficient touch-and-retract motions. In contrast, in this paper we propose to query information efficient sliding paths instead of only touch locations. This is realized by three components: A Gaussian process implicit surface model represents the shape and uncertainty of the object. A compliant task/force controller framework fuses the information of this GP model into the parameterization of its tasks, which enables the robot to slide over the unknown object safely and robustly. Thirdly, we develop two strategies to solve the proposed active path querying learning problem. Sliding along those query paths not only creates more dense data than touch probing, but additionally greatly reduces the uncertainty of the object. We demonstrate the effectiveness of our proposed framework both in simulation and on the PR2 robot platform. Furthermore, it is shown that our methodology can be extended to other learning tasks, such as finding a desired surface normal on an unknown object, e.g. for pushing. Danny Drieß, Peter Englert, Marc Toussaint	IROS	1