Achieving human-like dexterity is a longstanding challenge in robotics, in part due to the complexity of planning and control for contact-rich systems. In reinforcement learning (RL), one popular approach has been to use massively-parallelized, domain-randomized simulations to learn a policy offline over a vast array of contact conditions, allowing robust sim-to-real transfer. Inspired by recent advances in real-time parallel simulation, this work considers instead the viability of online planning methods for contact-rich manipulation by studying the well-known in-hand cube reorientation task. We propose a simple architecture that employs a sampling-based predictive controller and vision-based pose estimator to search for contact-rich control actions online. We conduct thorough experiments to assess the real-world performance of our method, architectural design choices, and key factors for robustness, demonstrating that our simple sampling-based approach achieves performance comparable to prior RL-based works. Supplemental material: https://caltech-amber.github.io/drop.

Robot learning has produced remarkably effective "black-box" controllers for complex tasks such as dynamic locomotion on humanoids. Yet ensuring dynamic safety, i.e., constraint satisfaction, remains challenging for such policies. Reinforcement learning (RL) embeds constraints heuristically through reward engineering, and adding or modifying constraints requires retraining. Model-based approaches, like control barrier functions (CBFs), enable runtime constraint specification with formal guarantees but require accurate dynamics models. This paper presents SHIELD, a layered safety framework that bridges this gap by: (1) training a generative, stochastic dynamics residual model using real-world data from hardware rollouts of the nominal controller, capturing system behavior and uncertainties; and (2) adding a safety layer on top of the nominal (learned locomotion) controller that leverages this model via a stochastic discrete-time CBF formulation enforcing safety constraints in probability. The result is a minimally-invasive safety layer that can be added to the existing autonomy stack to give probabilistic guarantees of safety that balance risk and performance. In hardware experiments on an Unitree G1 humanoid, SHIELD enables safe navigation (obstacle avoidance) through varied indoor and outdoor environments using a nominal (unknown) RL controller and onboard perception.

A key source of brittleness for robotic systems is the presence of model uncertainty and external disturbances. Most existing approaches to robust control either seek to bound the worst-case disturbance (which results in conservative behavior), or to learn a deterministic dynamics model (which is unable to capture uncertain dynamics or disturbances). This work proposes a different approach: training a state-conditioned generative model to represent the distribution of error residuals between the nominal dynamics and the actual system. In particular we introduce the Online Risk-Informed Optimization controller (ORIO), which uses Discrete-Time Control Barrier Functions, combined with a learned, generative disturbance model, to ensure the safety of the system up to some level of risk. We demonstrate our approach in simulations and hardware, and show that our method can learn a disturbance model that is accurate enough to enable risk-sensitive control of a quadrotor flying aggressively with an unmodelled slung load. We use a conditional variational autoencoder (CVAE) to learn a state-conditioned dynamics residual distribution, and find that the resulting controller can run at 100Hz on an embedded computer and exhibits less conservative behavior while retaining theoretical safety properties.

Conventional approaches to grasp planning re- quire perfect knowledge of an object’s pose and geometry. Uncertainties in these quantities induce uncertainties in the quality of planned grasps, which can lead to failure. Classically, grasp robustness refers to the ability to resist external disturbances after grasping an object. In contrast, this work studies robustness to intrinsic sources of uncertainty like object pose or geometry affecting grasp planning before execution. To do so, we develop a novel analytic theory of grasping that reasons about this intrinsic robustness by characterizing the effect of friction cone uncertainty on a grasp’s force closure status. We apply this result in two ways. First, we analyze the theoretical guarantees on intrinsic robustness of two grasp metrics in the literature, the classical Ferrari-Canny metric and more recent min-weight metric. We validate these results with hardware trials that compare grasps synthesized with and without robustness guarantees, showing a clear improvement in success rates. Second, we use our theory to develop a novel analytic notion of probabilistic force closure, which we show can generate unique, uncertainty-aware grasps in simulation.

We introduce a transformation of a Neural Radiance Field (NeRF) to an equivalent Poisson Point Process (PPP). This PPP transformation allows for rigorous quantification of uncertainty in NeRFs, in particular, for computing collision probabilities for a robot navigating through a NeRF environment. The PPP is a generalization of a probabilistic occupancy grid to the continuous volume and is fundamental to the volumetric ray-tracing model underlying radiance fields. Building upon this PPP representation, we present a chance-constrained trajectory optimization method for safe robot navigation in NeRFs. Our method relies on a voxel representation called the Probabilistic Unsafe Robot Region (PURR) that spatially fuses the chance constraint with the NeRF model to facilitate fast trajectory optimization. We then combine a graph-based search with a spline-based trajectory optimization to yield robot trajectories through the NeRF that are guaranteed to satisfy a user-specific collision probability. We validate our chance constrained planning method through simulations and hardware experiments, showing superior performance compared to prior works on trajectory planning in NeRF environments. Our code can be found athttps://github.com/chengine/catnips.

Many approaches to grasp synthesis optimize analytic quality metrics that measure grasp robustness based on finger placements and local surface geometry. However, generating feasible dexterous grasps by optimizing these metrics is slow, often taking minutes. To address this issue, this paper presents FRoGGeR: a method that quickly generates robust precision grasps using the min-weight metric, a novel, almost-everywhere differentiable approximation of the classical$\epsilon$grasp metric. The min-weight metric is simple and interpretable, provides a reasonable measure of grasp robustness, and admits numerically efficient gradients for smooth optimization. We leverage these properties to rapidly synthesize collision-free robust grasps-typically in less than a second. FRoGGeR can refine the candidate grasps generated by other methods (heuristic, data-driven, etc.) and is compatible with many object representations (SDFs, meshes, etc.), We study FRoGGeR's performance on over 40 objects drawn from the YCB dataset, outperforming a competitive baseline in computation time, feasibility rate of grasp synthesis, and picking success in simulation. We conclude that FRoGGeR is fast: it has a median synthesis time of 0.834s over hundreds of experiments.

Dexterous manipulation tasks often require contact switching, where fingers make and break contact with the object. We propose a method that plans trajectories for dexterous manipulation tasks involving contact switching using contact-implicit trajectory optimization (CITO) augmented with a high-level discrete contact sequence planner. We first use the high-level planner to find a sequence of finger contact switches given a desired object trajectory. With this contact sequence plan, we impose additional constraints in the CITO problem. We show that our method finds trajectories approximately 7 times faster than a general CITO baseline for a four-finger planar manipulation scenario. Furthermore, when executing the planned trajectories in a full dynamics simulator, we are able to more closely track the object pose trajectories planned by our method than those planned by the baselines.

In this work, we consider a group of robots working together to manipulate a rigid object to track a desired trajectory in$\text{SE}(3)$. The robots do not know the mass or friction properties of the object, or where they are attached to the object. They can, however, access a common state measurement, either from one robot broadcasting its measurements to the team, or by all robots communicating and averaging their state measurements to estimate the state of their centroid. To solve this problem, we propose a decentralized adaptive control scheme wherein each agent maintains and adapts its own estimate of the object parameters in order to track a reference trajectory. We present an analysis of the controller’s behavior, and show that all closed-loop signals remain bounded, and that the system trajectory will almost always (except for initial conditions on a set of measure zero) converge to the desired trajectory. We study the proposed controller’s performance using numerical simulations of a manipulation task in 3-D, as well as hardware experiments which demonstrate our algorithm on a planar manipulation task. These studies, taken together, demonstrate the effectiveness of the proposed controller even in the presence of numerous unmodeled effects, such as discretization errors and complex frictional interactions.

Multi-robot assembly has the potential to greatly reduce the cost and risk associated with the fabrication of large structures. Using teams of robots to perform assembly offers numerous advantages such as parallelism, robustness to single-agent failures, and flexibility in scheduling and task assignment. However, while previous work on multi-robot assembly focuses on generating feasible assembly plans and decentralized control strategies, we instead study the problem of planning optimal assembly sequences.To this end, we pose the problem of multi-robot assembly as a discrete optimization, specifically an integer linear program (ILP) or quadratic program (IQP), which aims to minimize the time to complete the assembly, or to minimize the distance traveled. We develop a model of multi-robot assembly that captures both geometric constraints and actuation constraints inherent to the problem. While the ILP and IQP can be solved exactly using commercial optimization software in a substantial amount of time, we also propose heuristic strategies which can be quickly computed, and can scale to structures of reasonable size. We also verify our methods empirically by comparing their performance on a variety of test structures.

This paper presents a design for a decentralized adaptive controller that allows a team of agents to manipulate a common payload in $\mathbb{R}^{2}$ or $\mathbb{R}^{3}$. The controller requires no communication between agents and requires no a priori knowledge of agent positions or payload properties. The agents can control the payload to track a reference trajectory in linear and angular velocity with center-of-mass measurements, in angular velocity using only local measurements and a common frame, and can stabilize its rotation with only local measurements. The controller is designed via a Lyapunov-style analysis and has proven stability and convergence. The controller is validated in simulation and experimentally with four robots manipulating an object in the plane.

Robots performing manipulation tasks must operate under uncertainty about both their pose and the dynamics of the system. In order to remain robust to modeling error and shifts in payload dynamics, agents must simultaneously perform estimation and control tasks. However, the optimal estimation actions are often not the optimal actions for accomplishing the control tasks, and thus agents trade between exploration and exploitation. This work frames the problem as a Bayes-adaptive Markov decision process and solves it online using Monte Carlo tree search and an extended Kalman filter to handle Gaussian process noise and parameter uncertainty in a continuous space. MCTS selects control actions to reduce model uncertainty and reach the goal state nearly optimally. Certainty equivalent model predictive control is used as a benchmark to compare performance in simulations with varying process noise and parameter uncertainty.