The generate-filter-refine (iterative paradigm) based on large language models (LLMs) has achieved progress in reasoning, programming, and program discovery in AI+Science. However, the effectiveness of search depends on where to search, namely, how to encode the domain prior into an operationally structured hypothesis space. To this end, this paper proposes a compact formal theory that describes and measures LLM-assisted iterative search guided by domain priors. We represent an agent as a fuzzy relation operator on inputs and outputs to capture feasible transitions; the agent is thereby constrained by a fixed safety envelope. To describe multi-step reasoning/search, we weight all reachable paths by a single continuation parameter and sum them to obtain a coverage generating function; this induces a measure of reachability difficulty; and it provides a geometric interpretation of search on the graph induced by the safety envelope. We further provide the simplest testable inferences and validate them via two instantiation. This theory offers a workable language and operational tools to measure agents and their search spaces, proposing a systematic formal description of iterative search constructed by LLMs.

A sensor-fused wearable assistance prototype for upper-limb function (triceps brachii and extensor pollicis brevis) is presented. The device integrates surface electromyography (sEMG), an inertial measurement unit (IMU), and flex/force sensors on an M5StickC plus an ESP32-S3 compute hub. Signals are band-pass and notch filtered; features (RMS, MAV, zero-crossings, and 4-12 Hz tremor-band power) are computed in 250 ms windows and fed to an INT8 TensorFlow Lite Micro model. Control commands are bounded by a control-barrier-function safety envelope and delivered within game-based tasks with lightweight personalization. In a pilot technical feasibility evaluation with healthy volunteers (n = 12) performing three ADL-oriented tasks, tremor prominence decreased (Delta TI = -0.092, 95% CI [-0.102, -0.079]), range of motion increased (+12.65%, 95% CI [+8.43, +13.89]), repetitions rose (+2.99 min^-1, 95% CI [+2.61, +3.35]), and the EMG median-frequency slope became less negative (Delta = +0.100 Hz/min, 95% CI [+0.083, +0.127]). The sensing-to-assist loop ran at 100 Hz with 8.7 ms median on-device latency, 100% session completion, and 0 device-related adverse events. These results demonstrate technical feasibility of embedded, sensor-fused assistance for upper-limb function; formal patient studies under IRB oversight are planned.

We present an adaptive scheduler for a single differencing engine (SmartDiff) with two execution modes: (i) in-memory threads and (ii) Dask based parallelism. The scheduler continuously tunes batch size and worker/thread count within fixed CPU and memory budgets to minimize p95 latency. A lightweight preflight profiler estimates bytes/row and I/O rate; an online cost/memory model prunes unsafe actions; and a guarded hill-climb policy favors lower latency with backpressure and straggler mitigation. Backend selection is gated by a conservative working-set estimate so that in-memory execution is chosen when safe, otherwise Dask is used. Across synthetic and public tabular benchmarks, the scheduler reduces p95 latency by 23 to 28 percent versus a tuned warm-up heuristic (and by 35 to 40 percent versus fixed grid baselines), while lowering peak memory by 16 to 22 percent (25 to 32 percent vs. fixed) with zero OOMs and comparable throughput.

Ensuring safety in autonomous driving requires precise, real-time risk assessment and adaptive behavior. Prior work on risk estimation either outputs coarse, global scene-level metrics lacking interpretability, proposes indicators without concrete integration into autonomous systems, or focuses narrowly on specific driving scenarios. We introduce the Context-aware Risk Index (CRI), a light-weight modular framework that quantifies directional risks based on object kinematics and spatial relationships, dynamically adjusting control commands in real time. CRI employs direction-aware spatial partitioning within a dynamic safety envelope using Responsibility-Sensitive Safety (RSS) principles, a hybrid probabilistic-max fusion strategy for risk aggregation, and an adaptive control policy for real-time behavior modulation. We evaluate CRI on the Bench2Drive benchmark comprising 220 safety-critical scenarios using a state-of-the-art end-to-end model Transfuser++ on challenging routes. Our collision-rate metrics show a 19\% reduction (p = 0.003) in vehicle collisions per failed route, a 20\% reduction (p = 0.004) in collisions per kilometer, a 17\% increase (p = 0.016) in composed driving score, and a statistically significant reduction in penalty scores (p = 0.013) with very low overhead (3.6 ms per decision cycle). These results demonstrate that CRI substantially improves safety and robustness in complex, risk-intensive environments while maintaining modularity and low runtime overhead.

Real-world accidents in learning-enabled CPS frequently occur in challenging corner cases. During the training of deep reinforcement learning (DRL) policy, the standard setup for training conditions is either fixed at a single initial condition or uniformly sampled from the admissible state space. This setup often overlooks the challenging but safety-critical corner cases. To bridge this gap, this paper proposes a physics-model-guided worst-case sampling strategy for training safe policies that can handle safety-critical cases toward guaranteed safety. Furthermore, we integrate the proposed worst-case sampling strategy into the physics-regulated deep reinforcement learning (Phy-DRL) framework to build a more data-efficient and safe learning algorithm for safety-critical CPS. We validate the proposed training strategy with Phy-DRL through extensive experiments on a simulated cart-pole system, a 2D quadrotor, a simulated and a real quadruped robot, showing remarkably improved sampling efficiency to learn more robust safe policies.

This paper proposes the SeC-Learning Machine: Simplex-enabled safe continual learning for safety-critical autonomous systems. The SeC-learning machine is built on Simplex logic (that is, ``using simplicity to control complexity'') and physics-regulated deep reinforcement learning (Phy-DRL). The SeC-learning machine thus constitutes HP (high performance)-Student, HA (high assurance)-Teacher, and Coordinator. Specifically, the HP-Student is a pre-trained high-performance but not fully verified Phy-DRL, continuing to learn in a real plant to tune the action policy to be safe. In contrast, the HA-Teacher is a mission-reduced, physics-model-based, and verified design. As a complementary, HA-Teacher has two missions: backing up safety and correcting unsafe learning. The Coordinator triggers the interaction and the switch between HP-Student and HA-Teacher. Powered by the three interactive components, the SeC-learning machine can i) assure lifetime safety (i.e., safety guarantee in any continual-learning stage, regardless of HP-Student's success or convergence), ii) address the Sim2Real gap, and iii) learn to tolerate unknown unknowns in real plants. The experiments on a cart-pole system and a real quadruped robot demonstrate the distinguished features of the SeC-learning machine, compared with continual learning built on state-of-the-art safe DRL frameworks with approaches to addressing the Sim2Real gap.

Safety architectures play a crucial role in the safety assurance of automated driving vehicles (ADVs). They can be used as safety envelopes of black-box ADV controllers, and for graceful degradation from one ODD to another. Building on our previous work on the formalization of responsibility-sensitive safety (RSS), we introduce a novel program logic that accommodates assume-guarantee reasoning and fallback-like constructs. This allows us to formally define and prove the safety of existing and novel safety architectures. We apply the logic to a pull over scenario and experimentally evaluate the resulting safety architecture.

A new generation of increasingly autonomous and self-learning cyber-physical systems (CPS) is being developed for control applications in the real world. These systems are AI-based in that they leverage techniques from the field of Artificial intelligence (AI) to flexibly cope with imprecision, inconsistency, incompleteness, to have an inherent ability to learn from experience, and to adapt according to changing and even unforeseen situations. This extra flexibility of AI systems makes it harder to predict their behavior. Moreover, AI systems usually are safety-critical in that they may be causing real harm in (and to) the real world. Consequently, the central question regarding the development of such systems is how to handle or even overcome this basic dichotomy between unpredictable and safe behavior of AI systems. In other words, how can we best construct systems that exploit AI techniques, without incurring the frailties of "AI-like" behavior?

To safely navigate unknown environments, robots must accurately perceive dynamic obstacles. Instead of directly measuring the scene depth with a LiDAR sensor, we explore the use of a much cheaper and higher resolution sensor: programmable light curtains. Light curtains are controllable depth sensors that sense only along a surface that a user selects. We use light curtains to estimate the safety envelope of a scene: a hypothetical surface that separates the robot from all obstacles. We show that generating light curtains that sense random locations (from a particular distribution) can quickly discover the safety envelope for scenes with unknown objects. Importantly, we produce theoretical safety guarantees on the probability of detecting an obstacle using random curtains. We combine random curtains with a machine learning based model that forecasts and tracks the motion of the safety envelope efficiently. Our method accurately estimates safety envelopes while providing probabilistic safety guarantees that can be used to certify the efficacy of a robot perception system to detect and avoid dynamic obstacles. We evaluate our approach in a simulated urban driving environment and a real-world environment with moving pedestrians using a light curtain device and show that we can estimate safety envelopes efficiently and effectively. Project website: https://siddancha.github.io/projects/active-safety-envelopes-with-guarantees

Balancing safety and efficiency when planning in dense traffic is challenging. Interactive behavior planners incorporate prediction uncertainty and interactivity inherent to these traffic situations. Yet, their use of single-objective optimality impedes interpretability of the resulting safety goal. Safety envelopes which restrict the allowed planning region yield interpretable safety under the presence of behavior uncertainty, yet, they sacrifice efficiency in dense traffic due to conservative driving. Studies show that humans balance safety and efficiency in dense traffic by accepting a probabilistic risk of violating the safety envelope. In this work, we adopt this safety objective for interactive planning. Specifically, we formalize this safety objective, present the Risk-Constrained Robust Stochastic Bayesian Game modeling interactive decisions satisfying a maximum risk of violating a safety envelope under uncertainty of other traffic participants' behavior and solve it using our variant of Multi-Agent Monte Carlo Tree Search. We demonstrate in simulation that our approach outperforms baselines approaches, and by reaching the specified violation risk level over driven simulation time, provides an interpretable and tunable safety objective for interactive planning.