Search for: All records

Latent safety filters extend Hamilton-Jacobi (HJ) reachability to operate on latent state representations and dynamics learned directly from high-dimensional observations, enabling safe visuomotor control under hard-to-model constraints. However, existing methods implement “leastrestrictive” filtering that discretely switch between nominal and safety policies, potentially undermining the task performance that makes modern visuomotor policies valuable. While reachability value functions can, in principle, be adapted to be control barrier functions (CBFs) for smooth optimization-based filtering, we theoretically and empirically show that current latentspace learning methods produce fundamentally incompatible value functions. We identify two sources of incompatibility: First, in HJ reachability, failures are encoded via a “margin function” in latent space, whose sign indicates whether or not a latent is in the constraint set. However, representing the margin function as a classifier yields saturated value functions that exhibit discontinuous jumps. We prove that the value function’s Lipschitz constant scales linearly with the margin function’s Lipschitz constant, revealing that smooth CBFs require smooth margins. Second, reinforcement learning (RL) approximations trained solely on safety policy data yield inaccurate value estimates for nominal policy actions, precisely where CBF filtering needs them. We propose the LatentCBF, which addresses both challenges through gradient penalties that lead to smooth margin functions without additional labeling, and a value-training procedure that mixes data from both nominal and safety policy distributions. Experiments on simulated benchmarks and hardware with a vision-based manipulation policy demonstrate that LatentCBF enables smooth safety filtering while doubling the task-completion rate over prior switching methods. 

Recent works have shown that foundational safe control methods, such as Hamilton-Jacobi (HJ) reachability analysis, can be applied in the latent space of world models. While this enables the synthesis of latent safety filters for hard-to-model vision-based tasks, they assume that the safety constraint is known a priori and remains fixed during deployment, limiting the safety filter's adaptability across scenarios. To address this, we propose constraint-parameterized latent safety filters that can adapt to user-specified safety constraints at runtime. Our key idea is to define safety constraints by conditioning on an encoding of an image that represents a constraint, using a latent-space similarity measure. The notion of similarity to failure is aligned in a principled way through conformal calibration, which controls how closely the system may approach the constraint representation. The parameterized safety filter is trained entirely within the world model's imagination, treating any image seen by the model as a potential test-time constraint, thereby enabling runtime adaptation to arbitrary safety constraints. In simulation and hardware experiments on vision-based control tasks with a Franka manipulator, we show that our method adapts at runtime by conditioning on the encoding of user-specified constraint images, without sacrificing performance. 

Recent advances in generative world models have enabled classical safe control methods, such as Hamilton-Jacobi (HJ) reachability, to generalize to complex robotic systems operating directly from high-dimensional sensor observations. However, obtaining comprehensive coverage of all safety-critical scenarios during world model training is extremely challenging. As a result, latent safety filters built on top of these models may miss novel hazards and even fail to prevent known ones, overconfidently misclassifying risky out-of-distribution (OOD) situations as safe. To address this, we introduce an uncertainty-aware latent safety filter that proactively steers robots away from both known and unseen failures. Our key idea is to use the world model's epistemic uncertainty as a proxy for identifying unseen potential hazards. We propose a principled method to detect OOD world model predictions by calibrating an uncertainty threshold via conformal prediction. By performing reachability analysis in an augmented state space-spanning both the latent representation and the epistemic uncertainty-we synthesize a latent safety filter that can reliably safeguard arbitrary policies from both known and unseen safety hazards. In simulation and hardware experiments on vision-based control tasks with a Franka manipulator, we show that our uncertainty-aware safety filter preemptively detects potential unsafe scenarios and reliably proposes safe, in-distribution actions. 

Hamilton-Jacobi (HJ) reachability is a rigorous mathematical framework that enables robots to simultaneously detect unsafe states and generate actions that prevent future failures. While in theory, HJ reachability can synthesize safe controllers for nonlinear systems and nonconvex constraints, in practice, it has been limited to hand-engineered collision avoidance constraints modeled via low-dimensional state-space representations and first-principles dynamics. In this work, our goal is to generalize safe robot controllers to prevent failures that are hard—if not impossible—to write down by hand, but can be intuitively identified from high-dimensional observations: for example, spilling the contents of a bag. We propose Latent Safety Filters, a latent-space generalization of HJ reachability that tractably operates directly on raw observation data (e.g., RGB images) to automatically computes safety-preserving actions without explicit recovery demonstrations by performing safety analysis in the latent embedding space of a generative world model. Our method leverages diverse robot observation-action data of varying quality (including successes, random exploration, and unsafe demonstrations) to learn a world model. Constraint specification is then transformed into a classification problem in the latent space of the learned world model. In simulation and hardware experiments, we compute an approximation of Latent Safety Filters to safeguard arbitrary policies (from imitation learned policies to direct teleoperation) from complex safety hazards, like preventing a Franka Research 3 manipulator from spilling the contents of a bag or toppling cluttered objects. 

With the advent of high repetition rate laser facilities, novel diagnostic tools compatible with these advanced specifications are required. This paper presents the design of an active gamma-ray spectrometer intended for these high repetition rate experiments, with particular emphasis on functionality within a PW level laser-plasma interaction chamber’s extreme conditions. The spectrometer uses stacked scintillators to accommodate a broad range of gamma-ray energies, demonstrating its adaptability for various experimental setups. In addition, it has been engineered to maintain compactness, electromagnetic pulse resistance, and ISO-5 cleanliness requirements while ensuring high sensitivity. The spectrometer has been tested in real conditions inside the PW-class level interaction chamber at the BELLA center, LBNL. The paper further details the calibration process, which utilizes a 60Co radioactive source, and describes the unfolding technique implemented through a stochastic minimization method. 

Abstract Radiation measurement relies on pulse detection, which can be performed using various configurations of high-speed analog-to-digital converters (ADCs) and field-programmable gate arrays (FPGAs). For optimal power consumption, design simplicity, system flexibility, and the availability of DSP slices, we consider the Radio Frequency System-on-Chip (RFSoC) to be a more suitable option than traditional setups. To this end, we have developed custom RFSoC-based electronics and verified its feasibility. The ADCs on RFSoC exhibit a flat frequency response of 1–125 MHz. The root-mean-square (RMS) noise level is 2.1 ADC without any digital signal processing. The digital signal processing improves the RMS noise level to 0.8 ADC (input equivalent 40 μV_rms). Baseline correction via digital signal processing can effectively prevent photomultiplier overshoot after a large pulse. Crosstalk between all channels is less than -55 dB. The measured data transfer speed can support up to 32 kHz trigger rates (corresponding to 750 Mbps). Overall, our RFSoC-based electronics are highly suitable for pulse detection, and after some modifications, they will be employed in the Kamioka Liquid Scintillator Anti-Neutrino Detector (KamLAND). 

Abstract The electron antineutrino flux limits are presented for the brightest gamma-ray burst (GRB) of all time, GRB221009A, over a range of 1.8–200 MeV using the Kamioka Liquid Scintillator Antineutrino Detector. Using multiple time windows ranging from minutes to days surrounding the event to search for electron antineutrinos coincident with the GRB, we set an upper limit on the flux under the assumption of several power-law neutrino source spectra, with power-law indices ranging from 1.5 to 3 in steps of 0.5. No excess was observed in any time windows ranging from seconds to days around the event trigger timeT₀. For a power-law index of 2 and a time window ofT₀ ±  500 s, a flux upper limit of 2.34  ×  10⁹cm⁻²was calculated. The limits are compared to the results presented by IceCube.