共找到 20 条结果
As one of the simplest non-prehensile manipulation skills, pushing has been widely studied as an effective means to rearrange objects. Existing approaches, however, typically rely on multi-step push plans composed of pre-defined pushing primitives with limited application scopes, which restrict their efficiency and versatility across different scenarios. In this work, we propose a unified pushing policy that incorporates a lightweight prompting mechanism into a flow matching policy to guide the generation of reactive, multimodal pushing actions. The visual prompt can be specified by a high-level planner, enabling the reuse of the pushing policy across a wide range of planning problems. Experimental results demonstrate that the proposed unified pushing policy not only outperforms existing baselines but also effectively serves as a low-level primitive within a VLM-guided planning framework to solve table-cleaning tasks efficiently.
In agricultural robotics, effective observation and localization of fruits present challenges due to occlusions caused by other parts of the tree, such as branches and leaves. These occlusions can result in false fruit localization or impede the robot from picking the fruit. The objective of this work is to push away branches that block the fruit's view to increase their visibility. Our setup consists of an RGB-D camera and a robot arm. First, we detect the occluded fruit in the RGB image and estimate its occluded part via a deep learning generative model in the depth space. The direction to push to clear the occlusions is determined using classic image processing techniques. We then introduce a 3D extension of the 2D Hough transform to detect straight line segments in the point cloud. This extension helps detect tree branches and identify the one mainly responsible for the occlusion. Finally, we clear the occlusion by pushing the branch with the robot arm. Our method uses a combination of deep learning for fruit appearance estimation, classic image processing for push direction determination, and 3D Hough transform for branch detection. We validate our perception methods through r
Push recovery during locomotion will facilitate the deployment of humanoid robots in human-centered environments. In this paper, we present a unified framework for walking control and push recovery for humanoid robots, leveraging the arms for push recovery while dynamically walking. The key innovation is to use the environment, such as walls, to facilitate push recovery by combining Single Rigid Body model predictive control (SRB-MPC) with Hybrid Linear Inverted Pendulum (HLIP) dynamics to enable robust locomotion, push detection, and recovery by utilizing the robot's arms to brace against such walls and dynamically adjusting the desired contact forces and stepping patterns. Extensive simulation results on a humanoid robot demonstrate improved perturbation rejection and tracking performance compared to HLIP alone, with the robot able to recover from pushes up to 100N for 0.2s while walking at commanded speeds up to 0.5m/s. Robustness is further validated in scenarios with angled walls and multi-directional pushes.
This paper presents push puppet networks, a novel Bayesian algorithm for structured pruning of large language models. The push puppet network learns a hierarchical function during training that can optimally determine specific network layers to keep for a given target size. By adding a small number of gating parameters via a hierarchical penalty function, the learned smooth function can allow for a network to be resized to very specific sizes without loading the full model into memory or requiring further post-training computation. The method compares favorably with existing approaches (SparseGPT, Wanda) at high pruning sizes (less than 50% of network structure) while realizing measurable speed-ups on conventional GPUs with PyTorch. Furthermore, push puppet networks can achieve significant speedups as candidates for speculative decoding.
We consider a wireless network where a source generates packets and forwards them to a network containing $n$ nodes. The nodes in the network use the asynchronous push, pull or push-pull gossip communication protocols to maintain the most recent updates from the source. We use the version age of information metric to quantify the freshness of information in the network. Prior to this work, only the push gossiping protocol has been studied for age of information analysis. In this paper, we use the stochastic hybrid systems (SHS) framework to obtain recursive equations for the expected version age of sets of nodes in the time limit. We then show that the pull and push-pull protocols can achieve constant version age, while it is already known that the push protocol can only achieve logarithmic version age. We then show that the push-pull protocol performs better than the push and the pull protocol. Finally, we carry out numerical simulations to evaluate these results.
\textsc{Cops and Robber} is a game played on graphs where a set of \textit{cops} aim to \textit{capture} the position of a single \textit{robber}. The main parameter of interest in this game is the \textit{cop number}, which is the minimum number of cops that are sufficient to guarantee the capture of the robber. In a directed graph $\overrightarrow{G}$, the \textit{push} operation on a vertex $v$ reverses the orientation of all arcs incident on $v$. We consider a variation of classical \textsc{Cops and Robber} on oriented graphs, where in its turn, each cop can either move to an out-neighbor of its current vertex or push some vertex of the graph, whereas, the robber can move to an adjacent vertex in its turn. [Das et al., CALDAM, 2023] introduced this variant and established that if $\overrightarrow{G}$ is an orientation of a subcubic graph, then one cop with push ability has a winning strategy. We extend these results to establish that if $\overrightarrow{G}$ is an orientation of a $3$-degenerate graph, or of a graph with maximum degree $4$, then one cop with push ability has a winning strategy.
Pushing is a motion primitive useful to handle objects that are too large, too heavy, or too cluttered to be grasped. It is at the core of much of robotic manipulation, in particular when physical interaction is involved. It seems reasonable then to wish for robots to understand how pushed objects move. In reality, however, robots often rely on approximations which yield models that are computable, but also restricted and inaccurate. Just how close are those models? How reasonable are the assumptions they are based on? To help answer these questions, and to get a better experimental understanding of pushing, we present a comprehensive and high-fidelity dataset of planar pushing experiments. The dataset contains timestamped poses of a circular pusher and a pushed object, as well as forces at the interaction.We vary the push interaction in 6 dimensions: surface material, shape of the pushed object, contact position, pushing direction, pushing speed, and pushing acceleration. An industrial robot automates the data capturing along precisely controlled position-velocity-acceleration trajectories of the pusher, which give dense samples of positions and forces of uniform quality. We finis
Pump-push-probe spectroscopy was employed for the exploration of charge-separation processes in organic photovoltaic blends as well as for proton-coupled electron-transfer (PCET) reactions in hydrogen-bonded complexes of tri-anisole-heptazine with substituted phenols in organic solvents. In the present work, the electron and proton transfer dynamics driven by a femtosecond pump pulse and a time-delayed femtosecond push pulse has been studied with ab initio on-the-fly nonadiabatic trajectory calculations for the hydrogen-bonded heptazine-H2O complex. While the dynamics following the pump pulse is dominated by ultrafast radiationless energy relaxation to the long-lived lowest singlet excited state (S1) of the heptazine chromophore with only minor PCET reactivity, the re-excitation of the transient S1 population by the push pulse results in a much higher PCET reaction probability. These results illustrate that pump-push excitation has the potential to unravel the individual electron and proton transfer processes of PCET reactions on femtosecond time scales.
Prolonged sitting is a health risk leading to metabolic and cardiovascular diseases. To combat this, various "nudging" strategies encourage stand-ups. Behavior change triggers use explicit prompts such as smartphone push notifications or light controls. However, comparisons of the effects of such interactions, discomfort, and user context have not yet been performed. The present study evaluated these methods in a mixed design experiment with 15 college students. Three intervention methods (none, push notifications, and light dimming) and three user task contexts (computer work, video calls, and reading) were tested. The frequency of standing up and comfort were assessed after each ten-minute session. Results showed that dimming resulted in slightly more breaks (1.4 \pm 1.55) than push notification (1.2 \pm 1.08), but caused discomfort for 66.7% of participants, compared to 20% for notification. The results were influenced by task context. Dimming was most effective during video calls and reading, while push notifications were more effective during computer work. These findings suggest adaptive nudging systems should tailor interventions based on context and individual preferences.
We present PushGen, an automated framework for generating high-quality push notifications comparable to human-crafted content. With the rise of generative models, there is growing interest in leveraging LLMs for push content generation. Although LLMs make content generation straightforward and cost-effective, maintaining stylistic control and reliable quality assessment remains challenging, as both directly impact user engagement. To address these issues, PushGen combines two key components: (1) a controllable category prompt technique to guide LLM outputs toward desired styles, and (2) a reward model that ranks and selects generated candidates. Extensive offline and online experiments demonstrate its effectiveness, which has been deployed in large-scale industrial applications, serving hundreds of millions of users daily.
Security plays a crucial role when it comes to planning large events such as concerts, sporting tournaments, pilgrims, or demonstrations. Monitoring and controlling pedestrian dynamics can prevent dangerous situations from occurring. However, little is known about the specific factors that contribute to harmful situations. For example, the individual response of a person to external forces in dense crowds is not well studied. In order to address this gap in knowledge, we conducted a series of experiments to examine how a push propagates through a row of people and how it affects the participants. We recorded 2D head trajectories and 3D motion capturing data. To ensure that different trials can be compared to one another, we measured the force at the impact. We find that that the propagation distance as well as the propagation speed of the push are mainly functions of the strength of the push and in particular the latter depends on the initial arm posture of the pushed participants. Our results can contribute to a deeper understanding of the microscopic causes of macroscopic phenomena in large groups, and can be applied to inform models of pedestrian dynamics or validate them, ultim
We prove that two pushing-blocks puzzles are intractable in 2D. One of our constructions improves an earlier result that established intractability in 3D [OS99] for a puzzle inspired by the game PushPush. The second construction answers a question we raised in [DDO00] for a variant we call Push-1. Both puzzles consist of unit square blocks on an integer lattice; all blocks are movable. An agent may push blocks (but never pull them) in attempting to move between given start and goal positions. In the PushPush version, the agent can only push one block at a time, and moreover when a block is pushed it slides the maximal extent of its free range. In the Push-1 version, the agent can only push one block one square at a time, the minimal extent---one square. Both NP-hardness proofs are by reduction from SAT, and rely on a common construction.
Viewpoint planning is an important task in any application where objects or scenes need to be viewed from different angles to achieve sufficient coverage. The mapping of confined spaces such as shelves is an especially challenging task since objects occlude each other and the scene can only be observed from the front, posing limitations on the possible viewpoints. In this paper, we propose a deep reinforcement learning framework that generates promising views aiming at reducing the map entropy. Additionally, the pipeline extends standard viewpoint planning by predicting adequate minimally invasive push actions to uncover occluded objects and increase the visible space. Using a 2.5D occupancy height map as state representation that can be efficiently updated, our system decides whether to plan a new viewpoint or perform a push. To learn feasible pushes, we use a neural network to sample push candidates on the map based on training data provided by human experts. As simulated and real-world experimental results with a robotic arm show, our system is able to significantly increase the mapped space compared to different baselines, while the executed push actions highly benefit the view
This paper develops a functional-analytic framework for approximating the push-forward induced by an analytic map from finitely many samples. Instead of working directly with the map, we study the push-forward on the space of locally analytic functionals and identify it, via the Fourier--Borel transform, with an operator on the space of entire functions of exponential type. This yields finite-dimensional approximations of the push-forward together with explicit error bounds expressed in terms of the smallest eigenvalues of certain Hankel moment matrices. Moreover, we obtain sample complexity bounds for the approximation from i.i.d.~sampled data. As a consequence, we show that linear algebraic operations on the finite-dimensional approximations can be used to reconstruct analytic vector fields from discrete trajectory data. In particular, we prove convergence of a data-driven method for recovering the vector field of an ordinary differential equation from finite-time flow map data under fairly general conditions.
In organic bulk heterojunction materials, charge delocalization has been proposed to play a vital role in the generation of free carriers by reducing the Coulomb attraction via an interfacial charge transfer exciton (CTX). Pump-push-probe (PPP) experiments produced evidence that the excess energy given by a push pulse enhances delocalization, thereby increasing photocurrent. However, previous studies have employed near-IR push pulses in the range 0.4-0.6 eV which is larger than the binding energy of a typical CTX. This raises the doubt that the push pulse may directly promote dissociation without involving delocalized states. Here, we perform PPP experiments with mid-IR push pulses at energies that are well below the binding energy of a CTX state (0.12-0.25 eV). We identify three types of CTX: delocalized, localized, and trapped. The excitation resides over multiple polymer chains in delocalized CTXs, while is restricted to a single chain (albeit maintaining a degree of intrachain delocalization) in localized CTXs. Trapped CTXs are instead completely localized. The pump pulse generates a hot delocalized CTX, which relaxes to a localized CTX, and eventually to trapped states. We fin
In robots, nonprehensile manipulation operations such as pushing are a useful way of moving large, heavy or unwieldy objects, moving multiple objects at once, or reducing uncertainty in the location or pose of objects. In this study, we propose a reactive and adaptive method for robotic pushing that uses rich feedback from a high-resolution optical tactile sensor to control push movements instead of relying on analytical or data-driven models of push interactions. Specifically, we use goal-driven tactile exploration to actively search for stable pushing configurations that cause the object to maintain its pose relative to the pusher while incrementally moving the pusher and object towards the target. We evaluate our method by pushing objects across planar and curved surfaces. For planar surfaces, we show that the method is accurate and robust to variations in initial contact position/angle, object shape and start position; for curved surfaces, the performance is degraded slightly. An immediate consequence of our work is that it shows that explicit models of push interactions might be sufficient but are not necessary for this type of task. It also raises the interesting question of
Personalized PageRank (PPR) is a popular node proximity metric in graph mining and network research. Given a graph G=(V,E) and a source node $s \in V$, a single-source PPR (SSPPR) query asks for the PPR value $\vpi(u)$ with respect to s, which represents the relative importance of node u in the context of the source node s. Among existing algorithms for SSPPR queries, LocalPush is a fundamental method which serves as a cornerstone for subsequent algorithms. In LocalPush, a push operation is a crucial primitive operation, which distributes the probability at a node u to ALL u's neighbors via the corresponding edges. Although this push operation works well on unweighted graphs, unfortunately, it can be rather inefficient on weighted graphs. In particular, on unbalanced weighted graphs where only a few of these edges take the majority of the total weight among them, the push operation would have to distribute insignificant probabilities along those edges which just take the minor weights, resulting in expensive overhead. To resolve this issue, we propose the EdgePush algorithm, a novel method for computing SSPPR queries on weighted graphs. EdgePush decomposes the aforementioned push o
Optical scattering force is used to reduce the loading time of single atoms to a cavity mode. Releasing a cold atomic ensemble above the resonator, we apply a push beam along the direction of gravity, offering fast atomic transport with narrow velocity distribution. We also observe in real time that, when the push beam is illuminated against gravity, single atoms slow down and even turn around in the mode, through the cavity-transmission measurement. Our method can be employed to make atom-cavity experiments more efficient.
Supernovae (SNe) enrich and energize the surrounding interstellar medium (ISM) and are a key mechanism in the galaxy feedback cycle. The heating of the ISM by supernova shocks, and its subsequent cooling is critical to future star formation. The cooling of the diffuse shock-heated ISM is dominated by ultraviolet (UV) emission lines. These cooling regions and interfaces have complex spatial structure on sub-parsec scales. Mapping this cooling process is essential to understanding the feedback cycle of galaxies, a major goal of the 2020 Astrophysics Decadal Survey. The Supernova remnants and Proxies for ReIonization Testbed Experiment (SPRITE) CubeSat Mission will house the first long-slit orbital spectrograph with sub-arcminute angular resolution covering far ultraviolet wavelengths (FUV; 1000 - 1750 angstroms) and access to the Lyman UV (lambda < 1216 angstroms). SPRITE aims to provide new insights into the stellar feedback that drives galaxy evolution by mapping key FUV emission lines at the interaction lines between supernova remnants (SNRs) and the ambient interstellar medium (ISM). SPRITE will also measure the ionizing escape from approximately 50 low-redshift (0.16 < z &
Modern web pages have complex structures comprised of up to hundreds of different resources, such as scripts and images. Server push is an HTTP/2 feature enabling servers to preemptively send resources to clients before they realize they need them to render a page. The objective is to reduce the amount of time the browser has to wait for data to be transferred, and consequently total page load time. Our goal in this work is to quantify how much server push can actually reduce web page load times. We approach the problem from both theoretical and experimental perspectives. We start by deriving an upper bound for the load time reduction afforded by server push. Then we proceed to actually evaluate an idealized push implementation on the Alexa Top 100 global sites, against a non-push HTTP/2 baseline. Our results show a linear relationship between latency and the benefit of server push. Moreover, pages with taller dependency trees tend to benefit the most from it.