arXiv subjects

Albert Wu

Publications and source records attributed to Albert Wu.

9 recordsLinked to original sources

arXiv2026

Safe and Steerable Geometric Motion Policies for Robotic Dexterous Manipulation

Robotic dexterous manipulation requires continuously reconciling objectives and constraints defined on heterogeneous geometric spaces: a robot controlled on a $\mathbb{R}^7$ configuration manifold may need to track end effector poses on $\mathrm{SE}(3)$ while satisfying obstacle avoidance margins in $\mathbb{R}$. We present Safe Pullback Bundle Dynamical Systems (SafePBDS), a geometrically consistent framework that computes optimal, certifiably safe configuration manifold accelerations from objectives and safety requirements on arbitrary task manifolds. SafePBDS builds on prior work that combines predefined task manifold dynamical systems to produce autonomous motion. Its first innovation is a pullback control barrier function construction, which converts task manifold safety conditions into linear constraints on configuration manifold accelerations. The second innovation is a task manifold action interface that allows a high-level policy to inject low dimensional residual motions; zero input recovers the autonomous behavior, while safety is preserved under arbitrary inputs. This lets high-level policies efficiently steer exploration while leaving precise motion to the autonomous behavior. We validate SafePBDS in simulation and on a 23-DOF Franka Panda-Allegro Hand platform. On dexterous grasping, SafePBDS achieves a $92.5\%$ success rate across 20 household objects and 120 trials. Using the action interface, the method can exclude any one of the four fingers during grasping via a one-dimensional action, achieving $94.4\%$ 3-finger grasp success across 3 objects and 36 trials. The efficient planning and safety guarantee of SafePBDS also enables the first model-based, fully actuated palm-down in-hand reorientation, exceeding $360^\circ$ of yaw rotation in both directions under varying object weight and wrist motion. Demo video and details: https://tml.stanford.edu/safe-pbds

cs.RO↗

arXiv2026

Test-Time Scaling Makes Overtraining Compute-Optimal

Modern LLMs scale at test-time, e.g. via repeated sampling, where inference cost grows with model size and the number of samples. This creates a trade-off that pretraining scaling laws, such as Chinchilla, do not address. We present Train-to-Test ($T^2$) scaling laws that jointly optimize model size, training tokens, and number of inference samples under fixed end-to-end budgets. $T^2$ modernizes pretraining scaling laws with pass@$k$ modeling used for test-time scaling, then jointly optimizes pretraining and test-time decisions. Forecasts from $T^2$ are robust over distinct modeling approaches: measuring joint scaling effect on the task loss and modeling impact on task accuracy. Across eight downstream tasks, we find that when accounting for inference cost, optimal pretraining decisions shift radically into the overtraining regime, well-outside of the range of standard pretraining scaling suites. We validate our results by pretraining heavily overtrained models in the optimal region that $T^2$ scaling forecasts, confirming their substantially stronger performance compared to pretraining scaling alone. Finally, as frontier LLMs are post-trained, we show that our findings survive the post-training stage, making $T^2$ scaling meaningful in modern deployments.

cs.LG↗

arXiv2024

In the Wild Ungraspable Object Picking with Bimanual Nonprehensile Manipulation

Picking diverse objects in the real world is a fundamental robotics skill. However, many objects in such settings are bulky, heavy, or irregularly shaped, making them ungraspable by conventional end effectors like suction grippers and parallel jaw grippers (PJGs). In this paper, we expand the range of pickable items without hardware modifications using bimanual nonprehensile manipulation. We focus on a grocery shopping scenario, where a bimanual mobile manipulator equipped with a suction gripper and a PJG is tasked with retrieving ungraspable items from tightly packed grocery shelves. From visual observations, our method first identifies optimal grasp points based on force closure and friction constraints. If the grasp points are occluded, a series of nonprehensile nudging motions are performed to clear the obstruction. A bimanual grasp utilizing contacts on the side of the end effectors is then executed to grasp the target item. In our replica grocery store, we achieved a 90% success rate over 102 trials in uncluttered scenes, and a 67% success rate over 45 trials in cluttered scenes. We also deployed our system to a real-world grocery store and successfully picked previously unseen items. Our results highlight the potential of bimanual nonprehensile manipulation for in-the-wild robotic picking tasks. A video summarizing this work can be found at youtu.be/g0hOrDuK8jM

cs.RO↗

arXiv2024

One-Shot Transfer of Long-Horizon Extrinsic Manipulation Through Contact Retargeting

Extrinsic manipulation, the use of environment contacts to achieve manipulation objectives, enables strategies that are otherwise impossible with a parallel jaw gripper. However, orchestrating a long-horizon sequence of contact interactions between the robot, object, and environment is notoriously challenging due to the scene diversity, large action space, and difficult contact dynamics. We observe that most extrinsic manipulation are combinations of short-horizon primitives, each of which depend strongly on initializing from a desirable contact configuration to succeed. Therefore, we propose to generalize one extrinsic manipulation trajectory to diverse objects and environments by retargeting contact requirements. We prepare a single library of robust short-horizon, goal-conditioned primitive policies, and design a framework to compose state constraints stemming from contacts specifications of each primitive. Given a test scene and a single demo prescribing the primitive sequence, our method enforces the state constraints on the test scene and find intermediate goal states using inverse kinematics. The goals are then tracked by the primitive policies. Using a 7+1 DoF robotic arm-gripper system, we achieved an overall success rate of 80.5% on hardware over 4 long-horizon extrinsic manipulation tasks, each with up to 4 primitives. Our experiments cover 10 objects and 6 environment configurations. We further show empirically that our method admits a wide range of demonstrations, and that contact retargeting is indeed the key to successfully combining primitives for long-horizon extrinsic manipulation. Code and additional details are available at stanford-tml.github.io/extrinsic-manipulation.

cs.RO↗

arXiv2023

Synthesize Dexterous Nonprehensile Pregrasp for Ungraspable Objects

Daily objects embedded in a contextual environment are often ungraspable initially. Whether it is a book sandwiched by other books on a fully packed bookshelf or a piece of paper lying flat on the desk, a series of nonprehensile pregrasp maneuvers is required to manipulate the object into a graspable state. Humans are proficient at utilizing environmental contacts to achieve manipulation tasks that are otherwise impossible, but synthesizing such nonprehensile pregrasp behaviors is challenging to existing methods. We present a novel method that combines graph search, optimal control, and a learning-based objective function to synthesize physically realistic and diverse nonprehensile pre-grasp motions that leverage the external contacts. Since the ``graspability'' of an object in context with its surrounding is difficult to define, we utilize a dataset of dexterous grasps to learn a metric which implicitly takes into account the exposed surface of the object and the finger tip locations. Our method can efficiently discover hand and object trajectories that are certified to be physically feasible by the simulation and kinematically achievable by the dexterous hand. We evaluate our method on eight challenging scenarios where nonprehensile pre-grasps are required to succeed. We also show that our method can be applied to unseen objects different from those in the training dataset. Finally, we report quantitative analyses on generalization and robustness of our method, as well as an ablation study.

cs.RO↗

arXiv2022

Learning Diverse and Physically Feasible Dexterous Grasps with Generative Model and Bilevel Optimization

To fully utilize the versatility of a multi-fingered dexterous robotic hand for executing diverse object grasps, one must consider the rich physical constraints introduced by hand-object interaction and object geometry. We propose an integrative approach of combining a generative model and a bilevel optimization (BO) to plan diverse grasp configurations on novel objects. First, a conditional variational autoencoder trained on merely six YCB objects predicts the finger placement directly from the object point cloud. The prediction is then used to seed a nonconvex BO that solves for a grasp configuration under collision, reachability, wrench closure, and friction constraints. Our method achieved an 86.7% success over 120 real world grasping trials on 20 household objects, including unseen and challenging geometries. Through quantitative empirical evaluations, we confirm that grasp configurations produced by our pipeline are indeed guaranteed to satisfy kinematic and dynamic constraints. A video summary of our results is available at youtu.be/9DTrImbN99I.

cs.RO↗

arXiv2022

Real-time Model Predictive Control and System Identification Using Differentiable Physics Simulation

Developing robot controllers in a simulated environment is advantageous but transferring the controllers to the target environment presents challenges, often referred to as the "sim-to-real gap". We present a method for continuous improvement of modeling and control after deploying the robot to a dynamically-changing target environment. We develop a differentiable physics simulation framework that performs online system identification and optimal control simultaneously, using the incoming observations from the target environment in real time. To ensure robust system identification against noisy observations, we devise an algorithm to assess the confidence of our estimated parameters, using numerical analysis of the dynamic equations. To ensure real-time optimal control, we adaptively schedule the optimization window in the future so that the optimized actions can be replenished faster than they are consumed, while staying as up-to-date with new sensor information as possible. The constant re-planning based on a constantly improved model allows the robot to swiftly adapt to the changing environment and utilize real-world data in the most sample-efficient way. Thanks to a fast differentiable physics simulator, the optimization for both system identification and control can be solved efficiently for robots operating in real time. We demonstrate our method on a set of examples in simulation and show that our results are favorable compared to baseline methods.

cs.RO↗

arXiv2022

Robust-RRT: Probabilistically-Complete Motion Planning for Uncertain Nonlinear Systems

Robust motion planning entails computing a global motion plan that is safe under all possible uncertainty realizations, be it in the system dynamics, the robot's initial position, or with respect to external disturbances. Current approaches for robust motion planning either lack theoretical guarantees, or make restrictive assumptions on the system dynamics and uncertainty distributions. In this paper, we address these limitations by proposing the robust rapidly-exploring random-tree (Robust-RRT) algorithm, which integrates forward reachability analysis directly into sampling-based control trajectory synthesis. We prove that Robust-RRT is probabilistically complete (PC) for nonlinear Lipschitz continuous dynamical systems with bounded uncertainty. In other words, Robust-RRT eventually finds a robust motion plan that is feasible under all possible uncertainty realizations assuming such a plan exists. Our analysis applies even to unstable systems that admit only short-horizon feasible plans; this is because we explicitly consider the time evolution of reachable sets along control trajectories. Thanks to the explicit consideration of time dependency in our analysis, PC applies to unstabilizable systems. To the best of our knowledge, this is the most general PC proof for robust sampling-based motion planning, in terms of the types of uncertainties and dynamical systems it can handle. Considering that an exact computation of reachable sets can be computationally expensive for some dynamical systems, we incorporate sampling-based reachability analysis into Robust-RRT and demonstrate our robust planner on nonlinear, underactuated, and hybrid systems.

cs.RO↗

arXiv2018

Estimating Causal Effects Under Interference Using Bayesian Generalized Propensity Scores

In most real-world systems units are interconnected and can be represented as networks consisting of nodes and edges. For instance, in social systems individuals can have social ties, family or financial relationships. In settings where some units are exposed to a treatment and its effect spills over connected units, estimating both the direct effect of the treatment and spillover effects presents several challenges. First, assumptions on the way and the extent to which spillover effects occur along the observed network are required. Second, in observational studies, where the treatment assignment is not under the control of the investigator, confounding and homophily are potential threats to the identification and estimation of causal effects on networks. Here, we make two structural assumptions: i) neighborhood interference, which assumes interference operates only through a function of the immediate neighbors' treatments ii) unconfoundedness of the individual and neighborhood treatment, which rules out the presence of unmeasured confounding variables, including those driving homophily. Under these assumptions we develop a new covariate-adjustment estimator for treatment and spillover effects in observational studies on networks. Estimation is based on a generalized propensity score that balances individual and neighborhood covariates across units under different levels of individual treatment and of exposure to neighbors' treatment. Adjustment for propensity score is performed using a penalized spline regression. Inference capitalizes on a three-step Bayesian procedure which allows to take into account the uncertainty in the propensity score estimation and avoiding model feedback. Finally, correlation of interacting units is taken into account using a community detection algorithm and incorporating random effects in the outcome model.

stat.ME↗