The mixed truck-drone delivery system has attracted increasing attention for its potential to optimize last-mile logistics. While the Flying Sidekick Traveling Salesman Problem (FSTSP) provides a foundation for modeling the truck-drone collaboration, it falls short of capturing real-world complexities by assuming a single truck-drone pair operating on a fully connected graph. We introduce the Multi-Agent FSTSP (MA-FSTSP), which extends FSTSP to handle multiple trucks, each carrying multiple drones operating over real road networks. Trucks must follow roads, while drones can fly directly between locations. To solve this NP-hard problem efficiently, we propose a novel three-phase algorithm that first partitions customers using a set-based distance heuristic, then computes initial truck routes via a Set TSP formulation, and finally optimizes drone deployment patterns by dynamic programming. Through extensive experiments on real-world road networks from Manhattan (1,024 nodes) and Boston (11,000 nodes), we demonstrate that our method achieves more than 30% cost reduction compared to existing approaches while scaling effectively to problems with 150 customers within a 20-minute computational time-bound.

Verifying safety of neural network control systems that use images as input is a difficult problem because, from a given system state, there is no known way to mathematically model what images are possible in the real-world. We build upon recent work that considers a surrogate verification approach, training a conditional generative adversarial network (cGAN) as an image generator in place of the real world. This setup enables set-based formal analysis of the closed-loop system, providing analysis beyond simulation and testing. While existing work is effective on small examples, excessive overapproximation both within a single control period (one-step error) and across multiple periods (multi-step error) limits its scalability. We propose approaches to overcome these errors. First, we address one-step error by composing the system's dynamics along with the cGAN and neural network controller, without losing the dependencies between input states and the control outputs as in the monotonic analysis of the system dynamics. Second, we reduce multi-step error by repeating the single-step composition, essentially unrolling multiple steps of the control loop into a large neural network. We then leverage existing network verification algorithms to compute accurate reachable sets for multiple steps, avoiding the accumulation of abstraction error at each step.We demonstrate the effectiveness of our approach in terms of both accuracy and scalability using two case studies. On the aircraft taxiing system, the converged reachable set is 175% larger using the prior baseline method compared with our proposed approach. On the emergency braking system, with 24x the number of image output variables from the cGAN, the baseline method fails to prove any states are safe, whereas our improvements enable set-based safety analysis.

Increased deployment of autonomous systems in fields like transportation and robotics have seen a corresponding increase in safety-critical failures. These failures can be difficult to model and debug due to the relative lack of data: compared to tens of thousands of examples from normal operations, we may have only seconds of data leading up to the failure. This scarcity makes it challenging to train generative models of rare failure events, as existing methods risk either overfitting to noise in the limited failure dataset or underfitting due to an overly strong prior. We address this challenge with CalNF, or calibrated normalizing flows, a self-regularized framework for posterior learning from limited data. CalNF achieves state-of-the-art performance on data-limited failure modeling and inverse problems and enables a first-of-a-kind case study into the root causes of the 2022 Southwest Airlines scheduling crisis.

While a lot of recent research focuses on enhancing the textual reasoning capabilities of Large Language Models (LLMs) by optimizing the multi-agent framework or reasoning chains, several benchmark tasks can be solved with 100\% success through direct coding, which is more scalable and avoids the computational overhead associated with textual iterating and searching. Textual reasoning has inherent limitations in solving tasks with challenges in math, logics, optimization, and searching, which is unlikely to be solved by simply scaling up the model and data size. The recently released OpenAI GPT Code Interpreter and multi-agent frameworks such as AutoGen have demonstrated remarkable proficiency of integrating code generation and execution to solve complex tasks using LLMs. However, based on our experiments on 7 existing popular methods for steering code/text generation in both single- and multi-turn settings with 14 tasks and 6 types of LLMs (including the new O1-preview), currently there is no optimal method to correctly steer LLMs to write code when needed. We discover some interesting patterns on when models use code vs. textual reasoning with the evolution to task complexity and model sizes, which even result in an astonishingly inverse scaling behavior. We also discover that results from LLM written code are not always better than using textual reasoning, even if the task could be solved through code. To mitigate the above issues, we propose three methods to better steer LLM code/text generation and achieve a notable improvement. The costs of token lengths and runtime are thoroughly discussed for all the methods. We believe the problem of steering LLM code/text generation is critical for future research and has much space for further improvement. Project Page, Datasets, and Codes are available at https://yongchao98.github.io/CodeSteer/.

While large language models (LLMs) have recently demonstrated strong potential in solving planning problems, there is a trade-off between flexibility and complexity. LLMs, as zero-shot planners themselves, are still not capable of directly generating valid plans for complex planning problems such as multi-constraint or long-horizon tasks. On the other hand, many frameworks aiming to solve complex planning problems often rely on task-specific preparatory efforts, such as task-specific in-context examples and pre-defined critics/verifiers, which limits their cross-task generalization capability. In this paper, we tackle these challenges by observing that the core of many planning problems lies in optimization problems: searching for the optimal solution (best plan) with goals subject to constraints (preconditions and effects of decisions). With LLMs' commonsense, reasoning, and programming capabilities, this opens up the possibilities of a universal LLM-based approach to planning problems. Inspired by this observation, we propose LLMFP, a general-purpose framework that leverages LLMs to capture key information from planning problems and formally formulate and solve them as optimization problems from scratch, with no task-specific examples needed. We apply LLMFP to 9 planning problems, ranging from multi-constraint decision making to multi-step planning problems, and demonstrate that LLMFP achieves on average 83.7\% and 86.8\% optimal rate across 9 tasks for GPT-4o and Claude 3.5 Sonnet, significantly outperforming the best baseline (direct planning with OpenAI o1-preview) with 37.6\% and 40.7\% improvements. We also validate components of LLMFP with ablation experiments and analyzed the underlying success and failure reasons.

Control policies that can achieve high task performance and satisfy safety constraints are desirable for any system, including multi-agent systems (MAS). One promising technique for ensuring the safety of MAS is distributed control barrier functions (CBF). However, it is difficult to design distributed CBF-based policies for MAS that can tackle unknown discrete-time dynamics, partial observability, changing neighborhoods, and input constraints, especially when a distributed high-performance nominal policy that can achieve the task is unavailable. To tackle these challenges, we propose **DGPPO**, a new framework that *simultaneously* learns both a *discrete* graph CBF which handles neighborhood changes and input constraints, and a distributed high-performance safe policy for MAS with unknown discrete-time dynamics. We empirically validate our claims on a suite of multi-agent tasks spanning three different simulation engines. The results suggest that, compared with existing methods, our DGPPO framework obtains policies that achieve high task performance (matching baselines that ignore the safety constraints), and high safety rates (matching the most conservative baselines), with a *constant* set of hyperparameters across all environments.

Existing methods fail to effectively steer Large Language Models (LLMs) between textual reasoning and code generation, leaving symbolic computing capabilities underutilized. We introduce CodeSteer, an effective method for guiding LLM code/text generation. We construct a comprehensive benchmark SymBench comprising 37 symbolic tasks with adjustable complexity and also synthesize datasets of 12k multi-turn guidance/generation trajectories and 5.5k guidance comparison pairs. We fine-tune the Llama-3-8B model with a newly designed multi-turn supervised fine-tuning (SFT) and direct preference optimization (DPO). The resulting model, CodeSteerLLM, augmented with the proposed symbolic and self-answer checkers, effectively guides the code/text generation of larger models. Augmenting GPT-4o with CodeSteer raises its average performance score from 53.3 to 86.4, even outperforming the existing best LLM OpenAI o1 (82.7), o1-preview (74.8), and DeepSeek R1 (76.8) across all 37 tasks (28 seen, 9 unseen). Trained for GPT-4o, CodeSteer demonstrates superior generalizability, providing an average 41.8 performance boost on Claude, Mistral, and GPT-3.5. CodeSteer-guided LLMs fully harness symbolic computing to maintain strong performance on highly complex tasks. Models, Datasets, and Codes are available at https://github.com/yongchao98/CodeSteer-v1.0 and https://huggingface.co/yongchao98.

Learning to solve complex tasks with signal temporal logic (STL) specifications is crucial to many real-world applications. However, most previous works only consider fixed or parametrized STL specifications due to the lack of a diverse STL dataset and encoders to effectively extract temporal logic information for downstream tasks. In this paper, we propose TeLoGraF, Temporal Logic Graph-encoded Flow, which utilizes Graph Neural Networks (GNN) encoder and flow-matching to learn solutions for general STL specifications. We identify four commonly used STL templates and collect a total of 200K specifications with paired demonstrations. We conduct extensive experiments in five simulation environments ranging from simple dynamical models in the 2D space to high-dimensional 7DoF Franka Panda robot arm and Ant quadruped navigation. Results show that our method outperforms other baselines in the STL satisfaction rate. Compared to classical STL planning algorithms, our approach is 10-100X faster in inference and can work on any system dynamics. Besides, we show our graph-encoding method’s capability to solve complex STLs and robustness to out-distribution STL specifications. Code is available at https://github.com/mengyuest/TeLoGraF

Multi-agent coordination is crucial for reliable multi-robot navigation in shared spaces such as automated warehouses. In regions of dense robot traffic, local coordination methods may fail to find a deadlock-free solution. In these scenarios, it is appropriate to let a central unit generate a global schedule that decides the passing order of robots. However, the runtime of such centralized coordination methods increases significantly with the problem scale. In this paper, we propose to leverage Graph Neural Network Variational Autoencoders (GNN-VAE) to solve the multi-agent coordination problem at scale faster than through centralized optimization. We formulate the coordination problem as a graph problem and collect ground truth data using a Mixed-Integer Linear Program (MILP) solver. During training, our learning framework encodes good quality solutions of the graph problem into a latent space. At inference time, solution samples are decoded from the sampled latent variables, and the lowest-cost sample is selected for coordination. By construction, our GNN-VAE framework returns solutions that always respect the constraints of the considered coordination problem. Numerical results show that our approach trained on small-scale problems can achieve high-quality solutions even for large-scale problems with 250 robots, being much faster than other baselines.

Recent works have shown great potential of Large Language Models (LLMs) in robot task and motion planning (TAMP). Current LLM approaches generate text- or code-based reasoning chains with sub-goals and action plans. However, they do not fully leverage LLMs’ symbolic computing and code generation capabilities. Many robot TAMP tasks involve complex optimization under multiple constraints, where pure textual reasoning is insufficient. While augmenting LLMs with predefined solvers and planners improves performance, it lacks generalization across tasks. Given LLMs’ growing coding proficiency, we enhance their TAMP capabilities by steering them to generate code as symbolic planners for optimization and constraint verification. Unlike prior work that uses code to interface with robot action modules or pre-designed planners, we steer LLMs to generate code as solvers, planners, and checkers for TAMP tasks requiring symbolic computing, while still leveraging textual reasoning to incorporate common sense. With a multi-round guidance and answer evolution framework, the proposed Code-as-Symbolic-Planner improves success rates by average 24.1% over best baseline methods across seven typical TAMP tasks and three popular LLMs. Code-as-Symbolic-Planner shows strong effectiveness and generalizability across discrete and continuous environments, 2D/3D simulations and real-world settings, as well as single- and multi-robot tasks with diverse requirements. See our project website†for prompts, videos, and code.

Yilun Hao, Yongchao Chen, Yang Zhang, Chuchu Fan. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

We address the problem of safe policy learning in multi-agent safety-critical autonomous systems. In such systems, it is necessary for each agent to meet the safety requirements at all times while also cooperating with other agents to accomplish the task. Toward this end, we propose a safe Hierarchical Multi-Agent Reinforcement Learning (HMARL) approach based on Control Barrier Functions (CBFs). Our proposed hierarchical approach decomposes the overall reinforcement learning problem into two levels –- learning joint cooperative behavior at the higher level and learning safe individual behavior at the lower or agent level conditioned on the high-level policy. Specifically, we propose a skill-based HMARL-CBF algorithm in which the higher-level problem involves learning a joint policy over the skills for all the agents and the lower-level problem involves learning policies to execute the skills safely with CBFs. We validate our approach on challenging environment scenarios whereby a large number of agents have to safely navigate through conflicting road networks. Compared with existing state-of-the-art methods, our approach significantly improves the safety achieving near perfect (within $5\%$) success/safety rate while also improving performance across all the environments.

The Time-Dependent Traveling Salesman Problem (TDTSP) extends the classical TSP by allowing dynamic edge weights that vary with departure time, reflecting real-world scenarios such as transportation networks, where travel times fluctuate due to congestion patterns. TDTSP violates symmetry, triangle inequality, and cyclic invariance properties of classical TSP, creating unique computational challenges. In this paper, we propose a neural model that extends MatNet from static asymmetric TSP to time-dependent settings by using an adjacency tensor to capture temporal variations, followed by a time-aware decoder. Our architecture addresses the unique challenge of asymmetry and triangle inequality violations that change dynamically over time. Beyond architectural innovations, our research reveals a critical evaluation insight: many practical TDTSP instances maintain the same optimal solution regardless of time-dependent edge weights. This exposes a fundamental limitation in current evaluation practices for TDTSP that rely solely on average travel time metrics across all instances. Such metrics fail to effectively distinguish between methods that genuinely capture temporal dynamics and those that merely perform well on static routing problems. Instead, we present extensive experiments on real-world datasets, evaluating our approach on both entire datasets and specifically filtered instances where temporal dependencies alter the optimal solution. Results show that our method achieves state-of-the-art average optimality gap on full instances and significant travel-time reduction on instances where time-aware routing saves time. These results demonstrate state-of-the-art ability to identify and exploit temporal dependencies, setting new standards for evaluating time-dependent routing problems.

Before deploying autonomous systems in safety-critical applications, we must be able to understand and verify the safety of these systems. For cases where the risk or cost of real-world testing is prohibitive, we propose a simulation-based framework for 1) predicting ways in which an autonomous system is likely to fail and 2) automatically adjusting the system's design and control policy to preemptively mitigate those failures. Existing tools for failure prediction struggle to search over high-dimensional environmental parameters, cannot efficiently handle end-to-end testing for systems with vision in the loop, and provide little guidance on how to mitigate failures once they are discovered. We approach this problem through the lens of approximate Bayesian inference, using differentiable simulation and rendering for efficient failure case prediction and repair (and providing a gradient-free version of our algorithm for cases where a differentiable simulator is not available). We include a theoretical and empirical evaluation of the tradeoffs between gradient-based and gradient-free methods, applying our approach to a range of robotics and control problems, including optimizing search patterns for robot swarms, UAV formation control, and robust network control. Compared to optimization-based falsification methods, our method predicts a more diverse, representative set of failure modes, and we find that our use of differentiable simulation yields solutions that have up to 10x lower cost and requires up to 2x fewer iterations to converge relative to gradient-free techniques. In hardware experiments, we find that repairing control policies using our method leads to a 5x robustness improvement.

Distributed, scalable, and safe control of large-scale multiagent systems is a challenging problem. In this article, we design a distributed framework for safe multiagent control in large-scale environments with obstacles, where a large number of agents are required to maintain safety using only local information and reach their goal locations. We introduce a new class of certificates, termed graph control barrier function (GCBF), which are based on the well-established control barrier function theory for safety guarantees and utilize a graph structure for scalable and generalizable distributed control of MAS. We develop a novel theoretical framework to prove the safety of an arbitrary-sized MAS with a single GCBF. We propose a new training framework GCBF+ that uses graph neural networks to parameterize a candidate GCBF and a distributed control policy. The proposed framework is distributed and is capable of taking point clouds from LiDAR, instead of actual state information, for real-world robotic applications. We illustrate the efficacy of the proposed method through various hardware experiments on a swarm of drones with objectives ranging from exchanging positions to docking on a moving target without collision. In addition, we perform extensive numerical experiments, where the number and density of agents, as well as the number of obstacles, increase. Empirical results show that in complex environments with agents with nonlinear dynamics (e.g., Crazyflie drones), GCBF+ outperforms the hand-crafted CBF-based method with the best performance by up to 20% for relatively small-scale MAS with up to 256 agents, and leading reinforcement learning (RL) methods by up to 40% for MAS with 1024 agents. Furthermore, the proposed method does not compromise on the performance, in terms of goal reaching, for achieving high safety rates, which is a common tradeoff in RL-based methods. Project website:https://mit-realm.github.io/gcbfplus/

Prompt optimization aims to find the best prompt to a large language model (LLM) for a given task. LLMs have been successfully used to help find and improve prompt candidates for single-step tasks. However, realistic tasks for agents are multi-step and introduce new challenges: (1) Prompt content is likely to be more extensive and complex, making it more difficult for LLMs to analyze errors, (2) the impact of an individual step is difficult to evaluate, and (3) different people may have varied preferences about task execution. While humans struggle to optimize prompts, they are good at providing feedback about LLM outputs; we therefore introduce a new LLM-driven discrete prompt optimization framework PROMST that incorporates human-designed feedback rules to automatically offer direct suggestions for improvement. We also use an extra learned heuristic model that predicts prompt performance to efficiently sample from prompt candidates. This approach significantly outperforms both human-engineered prompts and several other prompt optimization methods across 11 representative multi-step tasks (an average 10.6%-29.3% improvement to current best methods on five LLMs respectively). We believe our work can serve as a benchmark for automatic prompt optimization for LLM-driven multi-step tasks.

For effective human-robot interaction, robots need to understand, plan, and execute complex, long-horizon tasks described by natural language. Recent advances in large language models (LLMs) have shown promise for translating natural language into robot action sequences for complex tasks. However, existing approaches either translate the natural language directly into robot trajectories or factor the inference process by decomposing language into task sub-goals and relying on a motion planner to execute each sub-goal. When complex environmental and temporal constraints are involved, inference over planning tasks must be performed jointly with motion plans using traditional task-and-motion planning (TAMP) algorithms, making factorization into subgoals untenable. Rather than using LLMs to directly plan task sub-goals, we instead perform few-shot translation from natural language task descriptions to an intermediate task representation that can then be consumed by a TAMP algorithm to jointly solve the task and motion plan. To improve translation, we automatically detect and correct both syntactic and semantic errors via autoregressive re-prompting, resulting in significant improvements in task completion. We show that our approach outperforms several methods using LLMs as planners in complex task domains. See our project website§for prompts, videos, and code.

A flurry of recent work has demonstrated that pre-trained large language models (LLMs) can be effective task planners for a variety of single-robot tasks. The planning performance of LLMs is significantly improved via prompting techniques, such as in-context learning or re-prompting with state feedback, placing new importance on the token budget for the context window. An under-explored but natural next direction is to investigate LLMs as multi-robot task planners. However, long-horizon, heterogeneous multi-robot planning introduces new challenges of coordination while also pushing up against the limits of context window length. It is therefore critical to find token-efficient LLM planning frameworks that are also able to reason about the complexities of multi-robot coordination. In this work, we compare the task success rate and token efficiency of four multi-agent communication frameworks (centralized, decentralized, and two hybrid) as applied to four coordination-dependent multi-agent 2D task scenarios for increasing numbers of agents. We find that a hybrid framework achieves better task success rates across all four tasks and scales better to more agents. We further demonstrate the hybrid frameworks in 3D simulations where the vision-to-text problem and dynamical errors are considered. See our project website4for prompts, videos, and code.

Large-scale self-supervised models have recently revolutionized our ability to perform a variety of tasks within the vision and language domains. However, using such models for autonomous systems is challenging because of safety requirements: besides executing correct actions, an autonomous agent must also avoid the high cost and potentially fatal critical mistakes. Traditionally, self-supervised training mainly focuses on imitating previously observed behaviors, and the training demonstrations carry no notion of which behaviors should be explicitly avoided. In this work, we propose Control Barrier Transformer (ConBaT), an approach that learns safe behaviors from demonstrations in a self-supervised fashion. ConBaT is inspired by the concept of control barrier functions in control theory and uses a causal transformer that learns to predict safe robot actions autoregressively using a critic that requires minimal safety data labeling. During deployment, we employ a lightweight online optimization to find actions that ensure future states lie within the learned safe set. We apply our approach to different simulated control tasks and show that our method results in safer control policies compared to other classical and learning-based methods such as imitation learning, reinforcement learning, and model predictive control.

Control barrier functions (CBFs) have become popular as a safety filter to guarantee the safety of nonlinear dynamical systems for arbitrary inputs. However, it is difficult to construct functions that satisfy the CBF constraints for high relative degree systems with input constraints. To address these challenges, recent work has explored learning CBFs using neural networks via neural CBFs (NCBFs). However, such methods face difficulties when scaling to higher dimensional systems under input constraints. In this work, we first identify challenges that NCBFs face during training. Next, to address these challenges, we propose policy neural CBFs (PNCBFs), a method of constructing CBFs by learning the value function of a nominal policy, and show that the value function of the maximum-over-time cost is a CBF. We demonstrate the effectiveness of our method in simulation on a variety of systems ranging from toy linear systems to a jet aircraft with a 16-dimensional state space. Finally, we validate our approach on a two-agent quadcopter system on hardware under tight input constraints.

Sampling-based motion planning methods for manipulators in crowded environments often suffer from expensive collision checking and high sampling complexity, which make them difficult to use in real time. To address this issue, we propose a new generalizable control barrier function (CBF)based steering controller to reduce the number of samples needed in a sampling-based motion planner RRT. Our method combines the strength of CBF for real-time collision-avoidance control and RRT for long-horizon motion planning, by using CBF-induced neural controller (CBF-INC) to generate control signals that steer the system towards sampled configurations by RRT. CBF-INC is learned as Neural Networks and has two variants handling different inputs, respectively: state (signed distance) input and point-cloud input from LiDAR. In the latter case, we also study two different settings: fully and partially observed environmental information. Compared to manually crafted CBF which suffers from over-approximating robot geometry, CBF-INC can balance safety and goal-reaching better without being over-conservative. Given state-based input, our neural CBF-induced neural controller-enhanced RRT (CBFINC-RRT) can increase the success rate by 14% while reducing the number of nodes explored by 30%, compared with vanilla RRT on hard test cases. Given LiDAR input where vanilla RRT is not directly applicable, we demonstrate that our CBF-INCRRT can improve the success rate by 10%, compared with planning with other steering controllers. Our project page with supplementary material is at https://mit-realm.github.io/CBFINC-RRT-website/.

Current reinforcement-learning methods are unable to directly learn policies that solve the minimum cost reach-avoid problem to minimize cumulative costs subject to the constraints of reaching the goal and avoiding unsafe states, as the structure of this new optimization problem is incompatible with current methods. Instead, a surrogate problem is solved where all objectives are combined with a weighted sum. However, this surrogate objective results in suboptimal policies that do not directly minimize the cumulative cost. In this work, we propose RC-PPO, a reinforcement-learning-based method for solving the minimum-cost reach-avoid problem by using connections to Hamilton-Jacobi reachability. Empirical results demonstrate that RC-PPO learns policies with comparable goal-reaching rates to while achieving up to 57% lower cumulative costs compared to existing methods on a suite of minimum-cost reach-avoid benchmarks on the Mujoco simulator. The project page can be found at https://oswinso.xyz/rcppo.

Uncertainty in safety-critical cyber-physical systems can be modeled using a finite number of parameters or parameterized input signals. Given a system specification in Signal Temporal Logic (STL), we would like to verify that for all (infinite) values of the model parameters/input signals, the system satisfies its specification. Unfortunately, this problem is undecidable in general. Statistical model checking (SMC) offers a solution by providing guarantees on the correctness of CPS models by statistically reasoning on model simulations. We propose a new approach for statistical verification of CPS models for user-provided distribution on the model parameters. Our technique uses model simulations to learn surrogate models , and uses conformal inference to provide probabilistic guarantees on the satisfaction of a given STL property. Additionally, we can provide prediction intervals containing the quantitative satisfaction values of the given STL property for any user-specified confidence level. We compare this prediction interval with the interval we get using risk estimation procedures. We also propose a refinement procedure based on Gaussian Process (GP)-based surrogate models for obtaining fine-grained probabilistic guarantees over sub-regions in the parameter space. This in turn enables the CPS designer to choose assured validity domains in the parameter space for safety-critical applications. Finally, we demonstrate the efficacy of our technique on several CPS models.

Temporal Logic (TL) can be used to rigorously specify complex high-level specification for systems in many engineering applications.The translation between natural language (NL) and TL has been under-explored due to the lack of dataset and generalizable model across different application domains.In this paper, we propose an accurate and generalizable transformation framework of English instructions from NL to TL, exploring the use of Large Language Models (LLMs) at multiple stages.Our contributions are twofold.First, we develop a framework to create a dataset of NL-TL pairs combining LLMs and human annotation.We publish a dataset with 28K NL-TL pairs.Then, we finetune T5 models on the lifted versions (i.e., the specific Atomic Propositions (AP) are hidden) of the NL and TL.The enhanced generalizability originates from two aspects: 1) Usage of lifted NL-TL characterizes common logical structures, without constraints of specific domains.2) Application of LLMs in dataset creation largely enhances corpus richness.We test the generalization of trained models on five varied domains.To achieve full NL-TL transformation, we either combine the lifted model with AP recognition task or do the further finetuning on each specific domain.During the further finetuning, our model achieves higher accuracy (>95%) using only <10% training data, compared with the baseline sequence to sequence (Seq2Seq) model.12

Uncertainty is prevalent in robotics. Due to measurement noise and complex dynamics, we cannot estimate the exact system and environment state. Since conservative motion planners are not guaranteed to find a safe control strategy in a crowded, uncertain environment, we propose a density-based method. Our approach uses a neural network and the Liouville equation to learn the density evolution for a system with an uncertain initial state. We can plan for feasible and probably safe trajectories by applying a gradient-based optimization procedure to minimize the collision risk. We conduct motion planning experiments on simulated environments and environments generated from real-world data and outperform baseline methods such as model predictive control and nonlinear programming. While our method requires offline planning, the online run time is 100 times smaller compared to model predictive control. The code and supplementary material can be found at https://mit-realm.github.io/density_planner/.

To navigate complex environments, robots must increasingly use high-dimensional visual feedback (e.g. images) for control. However, relying on high-dimensional image data to make control decisions raises important questions; particularly, how might we prove the safety of a visual-feedback controller? Control barrier functions (CBFs) are powerful tools for certifying the safety of feedback controllers in the state-feedback setting, but CBFs have traditionally been poorly-suited to visual feedback control due to the need to predict future observations in order to evaluate the barrier function. In this work, we solve this issue by leveraging recent advances in neural radiance fields (NeRFs), which learn implicit representations of$\boldsymbol{3\mathrm{D}}$scenes and can render images from previously-unseen camera perspectives, to provide single-step visual foresight for a CBF-based controller, where the CBFs possess a discrete-time nature. This novel combination is able to filter out unsafe actions and intervene to preserve safety. We demonstrate the effect of our controller in real-time simulation experiments where it successfully prevents the robot from taking dangerous actions.

Learning-enabled control systems have demonstrated impressive empirical performance on challenging control problems in robotics, but this performance comes at the cost of reduced transparency and lack of guarantees on the safety or stability of the learned controllers. In recent years, new techniques have emerged to provide these guarantees by learning certificates alongside control policies—these certificates provide concise data-driven proofs that guarantee the safety and stability of the learned control system. These methods not only allow the user to verify the safety of a learned controller but also provide supervision during training, allowing safety and stability requirements to influence the training process itself. In this article, we provide a comprehensive survey of this rapidly developing field of certificate learning. We hope that this article will serve as an accessible introduction to the theory and practice of certificate learning, both to those who wish to apply these tools to practical robotics problems and to those who wish to dive more deeply into the theory of learning for control.

News events are often associated with quantities (e.g., the number of COVID-19 patients or the number of arrests in a protest), and it is often important to extract their type, time, and location from unstructured text in order to analyze these quantity events.This paper thus formulates the NLP problem of spatiotemporal quantity extraction, and proposes the first meta-framework for solving it.This meta-framework contains a formalism that decomposes the problem into several information extraction tasks, a shareable crowdsourcing pipeline, and transformer-based baseline models.We demonstrate the meta-framework in three domains-the COVID-19 pandemic, Black Lives Matter protests, and 2020 California wildfires-to show that the formalism is general and extensible, the crowdsourcing pipeline facilitates fast and high-quality data annotation, and the baseline system can handle spatiotemporal quantity extraction well enough to be practically useful.We release all resources for future research on this topic.1

Signal temporal logic (STL) provides a powerful, flexible framework for specifying complex autonomy tasks; however, existing methods for planning based on STL specifications have difficulty scaling to long-horizon tasks and are not robust to external disturbances. In this paper, we present an algorithm for finding robust plans that satisfy STL specifications. Our method alternates between local optimization and local falsification, using automatically differentiable temporal logic to iteratively optimize its plan in response to counterexamples found during the falsification process. We benchmark our counterexample-guided planning method against state-of-the-art planning methods on two long-horizon satellite rendezvous missions, showing that our method finds high-quality plans that satisfy STL specifications despite adversarial disturbances. We find that our method consistently finds plans that are robust to adversarial disturbances and requires less than half the time of competing methods. We provide an implementation of our planner at https://github.com/MIT-REALM/architect.

Learning-based methods have shown promising performance for accelerating motion planning, but mostly in the setting of static environments. For the more challenging problem of planning in dynamic environments, such as multi-arm assembly tasks and human-robot interaction, motion planners need to consider the trajectories of the dynamic obstacles and reason about temporal-spatial interactions in very large state spaces. We propose a GNN-based approach that uses temporal encoding and imitation learning with data aggregation for learning both the embeddings and the edge prioritization policies. Experiments show that the proposed methods can significantly accelerate online planning over state-of-the-art complete dynamic planning algorithms. The learned models can often reduce costly collision checking operations by more than 1000x, and thus accelerating planning by up to 95%, while achieving high success rates on hard instances as well.

We present a scalable and effective multi-agent safe motion planner that enables a group of agents to move to their desired locations while avoiding collisions with obstacles and other agents, with the presence of rich obstacles, high-dimensional, nonlinear, nonholonomic dynamics, actuation limits, and disturbances. We address this problem by finding a piecewise linear path for each agent such that the actual trajectories following these paths are guaranteed to satisfy the reach-and-avoid requirement. We show that the spatial tracking error of the actual trajectories of the controlled agents can be pre-computed for any qualified path that considers the minimum duration of each path segment due to actuation limits. Using these bounds, we find a collision-free path for each agent by solving Mixed Integer-Linear Programs and coordinate agents by using the priority-based search. We demonstrate our method by benchmarking in 2D and 3D scenarios with ground vehicles and quadrotors, respectively, and show improvements over the solving time and the solution quality compared to two state-of-the-art multi-agent motion planners.

Planning in hybrid systems with both discrete and continuous control variables is important for dealing with real-world applications such as extra-planetary exploration and multi-vehicle transportation systems. Meanwhile, generating high-quality solutions given certain hybrid planning specifications is crucial to building high-performance hybrid systems. However, since hybrid planning is challenging in general, most methods use greedy search that is guided by various heuristics, which is neither complete nor optimal and often falls into blind search towards an infinite-action plan. In this paper, we present a hybrid automaton planning formalism and propose an optimal approach that encodes this planning problem as a Mixed Integer Linear Program (MILP) by fixing the action number of automaton runs. We also show an extension of our approach for reasoning over temporally concurrent goals. By leveraging an efficient MILP optimizer, our method is able to generate provably optimal solutions for complex mixed discrete-continuous planning problems within a reasonable time. We use several case studies to demonstrate the extraordinary performance of our hybrid planning method and show that it outperforms a state-of-the-art hybrid planner, Scotty, in both efficiency and solution qualities.

We study constrained reinforcement learning (CRL) from a novel perspective by setting constraints directly on state density functions, rather than the value functions considered by previous works. State density has a clear physical and mathematical interpretation, and is able to express a wide variety of constraints such as resource limits and safety requirements. Density constraints can also avoid the time-consuming process of designing and tuning cost functions required by value function-based constraints to encode system specifications. We leverage the duality between density functions and Q functions to develop an effective algorithm to solve the density constrained RL problem optimally and the constrains are guaranteed to be satisfied. We prove that the proposed algorithm converges to a near-optimal solution with a bounded error even when the policy update is imperfect. We use a set of comprehensive experiments to demonstrate the advantages of our approach over state-of-the-art CRL methods, with a wide range of density constrained tasks as well as standard CRL benchmarks such as Safety-Gym.

Reactive and safe agent modellings are important for nowadays traffic simulator designs and safe planning applications. In this work, we proposed a reactive agent model which can ensure safety without comprising the original purposes, by learning only high-level decisions from expert data and a low level decentralized controller guided by the jointly learned decentralized barrier certificates. Empirical results show that our learned road user simulation models can achieve a significant improvement in safety comparing to state-of-the-art imitation learning and pure control-based methods, while being similar to human agents by having smaller error to the expert data. Moreover, our learned reactive agents are shown to generalize better to unseen traffic conditions, and react better to other road users and therefore can help understand challenging planning problems pragmatically.

We address the problem of synthesizing a controller for nonlinear systems with reach-avoid requirements. Our controller consists of a reference controller and a tracking controller which drives the actual trajectory to follow the reference trajectory. We identify a type of reference trajectory such that the tracking error between the actual trajectory of the closed-loop system and the reference trajectory can be bounded. Moreover, such a bound on the tracking error is independent of the reference trajectory. Using such bounds on the tracking error, we propose a method that can find a reference trajectory by solving a satisfiability problem over linear constraints. Our overall algorithm guarantees that the resulting controller can make sure every trajectory from the initial set of the system satisfies the given reach-avoid requirement. We also implement our technique in a tool FACTEST . We show that FACTEST can find controllers for four vehicle models (3–6 dimensional state space and 2–4 dimensional input space) across eight scenarios (with up to 22 obstacles), all with running time at the sub-second range.

We show that symmetry transformations and caching can enable scalable, and possibly unbounded, verification of multi-agent systems. Symmetry transformations map any solution of the system to another solution. We show that this property can be used to transform cached reachsets to compute new reachsets, for hybrid and multi-agent models. We develop a notion of a virtual system which defines symmetry transformations for a broad class of agent models that visit waypoint sequences. Using this notion of a virtual system, we present a prototype tool CacheReach that builds a cache of reachsets, in a way that is agnostic of the representation of the reachsets and the reachability analysis method used. Our experimental evaluation of CacheReach shows up to 64% savings in safety verification computation time on multi-agent systems with 3-dimensional linear and 4-dimensional nonlinear fixed-wing aircraft models following sequences of waypoints. These savings and our theoretical results illustrate the potential benefits of using symmetry-based caching in the safety verification of multi-agent systems.

We address the problem of synthesizing provably correct controllers for linear systems with reach-avoid specifications. Our solution uses a combination of an open-loop controller and a tracking controller, thereby reducing the problem to smaller tractable problems. We show that, once a tracking controller is fixed, the reachable states from an initial neighborhood, subject to any disturbance, can be over-approximated by a sequence of ellipsoids, with sizes that are independent of the open-loop controller. Hence, the open-loop controller can be synthesized independently to meet the reach-avoid specification for an initial neighborhood. Exploiting several techniques for tightening the over-approximations, we reduce the open-loop controller synthesis problem to satisfiability over quantifier-free linear real arithmetic. The overall synthesis algorithm, computes a tracking controller, and then iteratively covers the entire initial set to find open-loop controllers for initial neighborhoods. The algorithm is sound and, for a class of robust systems, is also complete. We present RealSyn , a tool implementing this synthesis algorithm, and we show that it scales to several high-dimensional systems with complex reach-avoid specifications. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

We present a demo of DryVR 2.0, a framework for verification and controller synthesis of cyber-physical systems composed of black-box simulators and white-box automata. For verification, DryVR 2.0 takes as input a black-box simulator, a white-box transition graph, a time bound and a safety specification. As output it generates over-approximations of the reachable states and returns "Safe" if the system meets the given bounded safety specification, or it returns "Unsafe" with a counter-example. For controller synthesis, DryVR 2.0 takes as input black-box simulator(s) and a reach-avoid specification, and uses RRTs to find a transition graph such that the combined system satisfies the given specification.

Simulation-driven verification can provide formal safety guarantees for otherwise intractable nonlinear and hybrid system models. A key step in simulation-driven algorithms is to compute the reach set overapproximations from a set of initial states through numerical simulations and sensitivity analysis. This article addresses this problem by providing algorithms for computing discrepancy functions as the upper bound on the sensitivity, that is, the rate at which trajectories starting from neighboring states converge or diverge. The algorithms rely on computing local bounds on matrix measures as the exponential change rate of the discrepancy function. We present two techniques to compute the matrix measures under different norms: regular Euclidean norm or Euclidean norm under coordinate transformation, such that the exponential rate of the discrepancy function, and therefore, the conservativeness of the overapproximation, is locally minimized. The proposed algorithms enable automatic reach set computations of general nonlinear systems and have been successfully used on several challenging benchmark models. All proposed algorithms for computing discrepancy functions give soundness and relative completeness of the overall simulation-driven safety-bounded verification algorithm. We present a series of experiments to illustrate the accuracy and performance of the algorithms.

Safety verification of embedded systems modeled as hybrid systems can be scaled up by employing simulation-guided reach set over-approximation techniques. Existing methods are either applicable to only restricted classes of systems, overly conservative, or computationally expensive. We present new techniques to compute a locally optimal bloating factor based on discrepancy functions, which allow construction of reach set over-approximations from simulation traces for general nonlinear systems. The discrepancy functions are critical for tools like C2E2 to verify bounded time safety properties for complex hybrid systems with nonlinear continuous dynamics. The new discrepancy function is computed using local bounds on a matrix measure under an optimal metric such that the exponential change rate of the discrepancy function is minimized. The new technique is less time consuming and less conservative than existing techniques and does not incur significant computational overhead. We demonstrate the effectiveness of our approach by comparing the performance of a prototype implementation with the state-of-the-art reachability analysis tool Flow

In this letter, we present a new algorithm for a single image super-resolution using the analysis sparse prior in thelαβ color space. Experimental results show that our algorithm outperforms other existing state-of-the-art methods. In addition, due to the high scalability of our algorithm, key modules of the proposed algorithm can be integrated with other super resolution algorithms.