Search

The rapid advancement of robotics necessitates robust tools for developing and testing safe control architectures in dynamic and uncertain environments. Ensuring safety and reliability in robotics, especially in safety-critical applications, is crucial, driving substantial industrial and academic efforts. In this context, we extend CBFkit, a Python/ROS2 toolbox, which now incorporates a planner using reach-avoid specifications as a cost function. This integration with the Model Predictive Path Integral (MPPI) controllers enables the toolbox to satisfy complex tasks while ensuring formal safety guarantees under various sources of uncertainty using Control Barrier Functions (CBFs). CBFkit is optimized for speed using JAX for automatic differentiation and jaxopt for quadratic program solving. The toolbox supports various robotic applications, including autonomous navigation, human-robot interaction, and multi-robot coordination. The toolbox also offers a comprehensive library of planner, controller, sensor, and estimator implementations. Through a series of examples, we demonstrate the enhanced capabilities of CBFkit in different robotic scenarios.

This paper introduces CBFKit, a Python/ROS toolbox for safe robotics planning and control under uncertainty. The toolbox provides a general framework for designing control barrier functions for mobility systems within both deterministic and stochastic environments. It can be connected to the ROS open-source robotics middleware, allowing for the setup of multi-robot applications, encoding of environments and maps, and integrations with predictive motion planning algorithms. Additionally, it offers multiple CBF variations and algorithms for robot control. The CBFKit is demonstrated on the Toyota Human Support Robot (HSR) in both simulation and in physical experiments., Comment: 8 pages

This paper introduces a model-based approach for training feedback controllers for an autonomous agent operating in a highly nonlinear (albeit deterministic) environment. We desire the trained policy to ensure that the agent satisfies specific task objectives and safety constraints, both expressed in Discrete-Time Signal Temporal Logic (DT-STL). One advantage for reformulation of a task via formal frameworks, like DT-STL, is that it permits quantitative satisfaction semantics. In other words, given a trajectory and a DT-STL formula, we can compute the {\em robustness}, which can be interpreted as an approximate signed distance between the trajectory and the set of trajectories satisfying the formula. We utilize feedback control, and we assume a feed forward neural network for learning the feedback controller. We show how this learning problem is similar to training recurrent neural networks (RNNs), where the number of recurrent units is proportional to the temporal horizon of the agent's task objectives. This poses a challenge: RNNs are susceptible to vanishing and exploding gradients, and na\"{i}ve gradient descent-based strategies to solve long-horizon task objectives thus suffer from the same problems. To tackle this challenge, we introduce a novel gradient approximation algorithm based on the idea of dropout or gradient sampling. One of the main contributions is the notion of {\em controller network dropout}, where we approximate the NN controller in several time-steps in the task horizon by the control input obtained using the controller in a previous training step. We show that our control synthesis methodology, can be quite helpful for stochastic gradient descent to converge with less numerical issues, enabling scalable backpropagation over long time horizons and trajectories over high dimensional state spaces.

Quadratic programs (QP) subject to multiple time-dependent control barrier function (CBF) based constraints have been used to design safety-critical controllers. However, ensuring the existence of a solution at all times to the QP subject to multiple CBF constraints is non-trivial. We quantify the feasible solution space of the QP in terms of its volume. We introduce a novel feasible space volume monitoring control barrier function that promotes compatibility of barrier functions and, hence, existence of a solution at all times. We show empirically that our approach not only enhances feasibility but also exhibits reduced sensitivity to changes in the hyperparameters such as gains of nominal controller. Finally, paired with a global planner, we evaluate our controller for navigation among humans in the AWS Hospital gazebo environment. The proposed controller is demonstrated to outperform the standard CBF-QP controller in maintaining feasibility.

In this work, we assess the theoretical limitations of determining guaranteed stability and accuracy of neural networks in classification tasks. We consider classical distribution-agnostic framework and algorithms minimising empirical risks and potentially subjected to some weights regularisation. We show that there is a large family of tasks for which computing and verifying ideal stable and accurate neural networks in the above settings is extremely challenging, if at all possible, even when such ideal solutions exist within the given class of neural architectures.

We propose a novel class of risk-aware control barrier functions (RA-CBFs) for the control of stochastic safety-critical systems. Leveraging a result from the stochastic level-crossing literature, we deviate from the martingale theory that is currently used in stochastic CBF techniques and prove that a RA-CBF based control synthesis confers a tighter upper bound on the probability of the system becoming unsafe within a finite time interval than existing approaches. We highlight the advantages of our proposed approach over the state-of-the-art via a comparative study on an mobile-robot example, and further demonstrate its viability on an autonomous vehicle highway merging problem in dense traffic., Comment: 7 pages, 4 figures, 5 tables, accepted at ICRA 2023

Signal Temporal Logic (STL) has become a popular tool for expressing formal requirements of Cyber-Physical Systems (CPS). The problem of verifying STL properties of neural network-controlled CPS remains a largely unexplored problem. In this paper, we present a model for the verification of Neural Network (NN) controllers for general STL specifications using a custom neural architecture where we map an STL formula into a feed-forward neural network with ReLU activation. In the case where both our plant model and the controller are ReLU-activated neural networks, we reduce the STL verification problem to reachability in ReLU neural networks. We also propose a new approach for neural network controllers with general activation functions; this approach is a sound and complete verification approach based on computing the Lipschitz constant of the closed-loop control system. We demonstrate the practical efficacy of our techniques on a number of examples of learning-enabled control systems.

In this work, we propose the model of timed partial orders (TPOs) for specifying workflow schedules, especially for modeling manufacturing processes. TPOs integrate partial orders over events in a workflow, specifying ``happens-before'' relations, with timing constraints specified using guards and resets on clocks -- an idea borrowed from timed-automata specifications. TPOs naturally allow us to capture event ordering, along with a restricted but useful class of timing relationships. Next, we consider the problem of mining TPO schedules from workflow logs, which include events along with their time stamps. We demonstrate a relationship between formulating TPOs and the graph-coloring problem, and present an algorithm for learning TPOs with correctness guarantees. We demonstrate our approach on synthetic datasets, including two datasets inspired by real-life applications of aircraft turnaround and gameplay videos of the Overcooked computer game. Our TPO mining algorithm can infer TPOs involving hundreds of events from thousands of data-points within a few seconds. We show that the resulting TPOs provide useful insights into the dependencies and timing constraints for workflows.

This study presents a novel approach for modeling and simulating human-vehicle interactions in order to examine the effects of automated driving systems (ADS) on driving performance and driver control workload. Existing driver-ADS interaction studies have relied on simulated or real-world human driver experiments that are limited in providing objective evaluation of the dynamic interactions and control workloads on the driver. Our approach leverages an integrated human model-based active driving system (HuMADS) to simulate the dynamic interaction between the driver model and the haptic-based ADS during a vehicle overtaking task. Two driver arm-steering models were developed for both tense and relaxed human driver conditions and validated against experimental data. We conducted a simulation study to evaluate the effects of three different haptic shared control conditions (based on the presence and type of control conflict) on overtaking task performance and driver workloads. We found that No Conflict shared control scenarios result in improved driving performance and reduced control workloads, while Conflict scenarios result in unsafe maneuvers and increased workloads. These findings, which are consistent with experimental studies, demonstrate the potential for our approach to improving future ADS design for safer driver assistance systems., Comment: 11 pages, 13 figures, submitted to the IEEE Transactions on Intelligent Vehicles

In this paper, we consider the problem of synthesizing a controller in the presence of uncertainty such that the resulting closed-loop system satisfies certain hard constraints while optimizing certain (soft) performance objectives. We assume that the hard constraints encoding safety or mission-critical task objectives are expressed using Signal Temporal Logic (STL), while performance is quantified using standard cost functions on system trajectories. In order to prioritize the satisfaction of the hard STL constraints, we utilize the framework of control barrier functions (CBFs) and algorithmically obtain CBFs for STL objectives. We assume that the controllers are modeled using neural networks (NNs) and provide an optimization algorithm to learn the optimal parameters for the NN controller that optimize the performance at a user-specified robustness margin for the safety specifications. We use the formalism of risk measures to evaluate the risk incurred by the trade-off between robustness margin of the system and its performance. We demonstrate the efficacy of our approach on well-known difficult examples for nonlinear control such as a quad-rotor and a unicycle, where the mission objectives for each system include hard timing constraints and safety objectives.

In this paper, we study Stochastic Control Barrier Functions (SCBFs) to enable the design of probabilistic safe real-time controllers in presence of uncertainties and based on noisy measurements. Our goal is to design controllers that bound the probability of a system failure in finite-time to a given desired value. To that end, we first estimate the system states from the noisy measurements using an Extended Kalman filter, and compute confidence intervals on the filtering errors. Then, we account for filtering errors and derive sufficient conditions on the control input based on the estimated states to bound the probability that the real states of the system enter an unsafe region within a finite time interval. We show that these sufficient conditions are linear constraints on the control input, and, hence, they can be used in tractable optimization problems to achieve safety, in addition to other properties like reachability, and stability. Our approach is evaluated using a simulation of a lane-changing scenario on a highway with dense traffic., Comment: CDC 2021, 60th Conference on Decision and Control, 7 pages

Safety is a critical concern for the next generation of autonomy that is likely to rely heavily on deep neural networks for perception and control. Formally verifying the safety and robustness of well-trained DNNs and learning-enabled systems under attacks, model uncertainties, and sensing errors is essential for safe autonomy. This research proposes a framework to repair unsafe DNNs in safety-critical systems with reachability analysis. The repair process is inspired by adversarial training which has demonstrated high effectiveness in improving the safety and robustness of DNNs. Different from traditional adversarial training approaches where adversarial examples are utilized from random attacks and may not be representative of all unsafe behaviors, our repair process uses reachability analysis to compute the exact unsafe regions and identify sufficiently representative examples to enhance the efficacy and efficiency of the adversarial training. The performance of our framework is evaluated on two types of benchmarks without safe models as references. One is a DNN controller for aircraft collision avoidance with access to training data. The other is a rocket lander where our framework can be seamlessly integrated with the well-known deep deterministic policy gradient (DDPG) reinforcement learning algorithm. The experimental results show that our framework can successfully repair all instances on multiple safety specifications with negligible performance degradation. In addition, to increase the computational and memory efficiency of the reachability analysis algorithm, we propose a depth-first-search algorithm that combines an existing exact analysis method with an over-approximation approach based on a new set representation. Experimental results show that our method achieves a five-fold improvement in runtime and a two-fold improvement in memory usage compared to exact analysis.

Deep convolutional neural networks have been widely employed as an effective technique to handle complex and practical problems. However, one of the fundamental problems is the lack of formal methods to analyze their behavior. To address this challenge, we propose an approach to compute the exact reachable sets of a network given an input domain, where the reachable set is represented by the face lattice structure. Besides the computation of reachable sets, our approach is also capable of backtracking to the input domain given an output reachable set. Therefore, a full analysis of a network's behavior can be realized. In addition, an approach for fast analysis is also introduced, which conducts fast computation of reachable sets by considering selected sensitive neurons in each layer. The exact pixel-level reachability analysis method is evaluated on a CNN for the CIFAR10 dataset and compared to related works. The fast analysis method is evaluated over a CNN CIFAR10 dataset and VGG16 architecture for the ImageNet dataset.

Sampling-based methods such as Rapidly-exploring Random Trees (RRTs) have been widely used for generating motion paths for autonomous mobile systems. In this work, we extend time-based RRTs with Control Barrier Functions (CBFs) to generate, safe motion plans in dynamic environments with many pedestrians. Our framework is based upon a human motion prediction model which is well suited for indoor narrow environments. We demonstrate our approach on a high-fidelity model of the Toyota Human Support Robot navigating in narrow corridors. We show in three scenarios that our proposed online method can navigate safely in the presence of moving agents with unknown dynamics.

Autonomous vehicles are complex systems that are challenging to test and debug. A requirements-driven approach to the development process can decrease the resources required to design and test these systems, while simultaneously increasing the reliability. We present a testing framework that uses signal temporal logic (STL), which is a precise and unambiguous requirements language. Our framework evaluates test cases against the STL formulae and additionally uses the requirements to automatically identify test cases that fail to satisfy the requirements. One of the key features of our tool is the support for machine learning (ML) components in the system design, such as deep neural networks. The framework allows evaluation of the control algorithms, including the ML components, and it also includes models of CCD camera, lidar, and radar sensors, as well as the vehicle environment. We use multiple methods to generate test cases, including covering arrays, which is an efficient method to search discrete variable spaces. The resulting test cases can be used to debug the controller design by identifying controller behaviors that do not satisfy requirements. The test cases can also enhance the testing phase of development by identifying critical corner cases that correspond to the limits of the system's allowed behaviors. We present STL requirements for an autonomous vehicle system, which capture both component-level and system-level behaviors. Additionally, we present three driving scenarios and demonstrate how our requirements-driven testing framework can be used to identify critical system behaviors, which can be used to support the development process., Comment: arXiv admin note: text overlap with arXiv:1804.06760

Pavement crack detection is a critical task for insuring road safety. Manual crack detection is extremely time-consuming. Therefore, an automatic road crack detection method is required to boost this progress. However, it remains a challenging task due to the intensity inhomogeneity of cracks and complexity of the background, e.g., the low contrast with surrounding pavements and possible shadows with similar intensity. Inspired by recent advances of deep learning in computer vision, we propose a novel network architecture, named Feature Pyramid and Hierarchical Boosting Network (FPHBN), for pavement crack detection. The proposed network integrates semantic information to low-level features for crack detection in a feature pyramid way. And, it balances the contribution of both easy and hard samples to loss by nested sample reweighting in a hierarchical way. To demonstrate the superiority and generality of the proposed method, we evaluate the proposed method on five crack datasets and compare it with state-of-the-art crack detection, edge detection, semantic segmentation methods. Extensive experiments show that the proposed method outperforms these state-of-the-art methods in terms of accuracy and generality.

This paper presents a technology for simple and computationally efficient improvements of a generic Artificial Intelligence (AI) system, including Multilayer and Deep Learning neural networks. The improvements are, in essence, small network ensembles constructed on top of the existing AI architectures. Theoretical foundations of the technology are based on Stochastic Separation Theorems and the ideas of the concentration of measure. We show that, subject to mild technical assumptions on statistical properties of internal signals in the original AI system, the technology enables instantaneous and computationally efficient removal of spurious and systematic errors with probability close to one on the datasets which are exponentially large in dimension. The method is illustrated with numerical examples and a case study of ten digits recognition from American Sign Language.

Freeway merging in congested traffic is a significant challenge toward fully automated driving. Merging vehicles need to decide not only how to merge into a spot, but also where to merge. We present a method for the freeway merging based on multi-policy decision making with a reinforcement learning method called {\em passive actor-critic} (pAC), which learns with less knowledge of the system and without active exploration. The method selects a merging spot candidate by using the state value learned with pAC. We evaluate our method using real traffic data. Our experiments show that pAC achieves 92\% success rate to merge into a freeway, which is comparable to human decision making., Comment: 6 pages, 5 figures. ICML Workshop on Machine Learning for Autonomous Vehicles

In many robotic applications, some aspects of the system dynamics can be modeled accurately while others are difficult to obtain or model. We present a novel reinforcement learning (RL) method for continuous state and action spaces that learns with partial knowledge of the system and without active exploration. It solves linearly-solvable Markov decision processes (L-MDPs), which are well suited for continuous state and action spaces, based on an actor-critic architecture. Compared to previous RL methods for L-MDPs and path integral methods which are model based, the actor-critic learning does not need a model of the uncontrolled dynamics and, importantly, transition noise levels; however, it requires knowing the control dynamics for the problem. We evaluate our method on two synthetic test problems, and one real-world problem in simulation and using real traffic data. Our experiments demonstrate improved learning and policy performance., Comment: 10 pages, 7 figures

Context in image is crucial for scene labeling while existing methods only exploit local context generated from a small surrounding area of an image patch or a pixel, by contrast long-range and global contextual information is ignored. To handle this issue, we in this work propose a novel approach for scene labeling by exploring multi-level contextual recurrent neural networks (ML-CRNNs). Specifically, we encode three kinds of contextual cues, i.e., local context, global context and image topic context in structural recurrent neural networks (RNNs) to model long-range local and global dependencies in image. In this way, our method is able to `see' the image in terms of both long-range local and holistic views, and make a more reliable inference for image labeling. Besides, we integrate the proposed contextual RNNs into hierarchical convolutional neural networks (CNNs), and exploit dependence relationships in multiple levels to provide rich spatial and semantic information. Moreover, we novelly adopt an attention model to effectively merge multiple levels and show that it outperforms average- or max-pooling fusion strategies. Extensive experiments demonstrate that the proposed approach achieves new state-of-the-art results on the CamVid, SiftFlow and Stanford-background datasets., Comment: 8 pages, 8 figures

In this work we discuss the problem of selecting suitable approximators from families of parameterized elementary functions that are known to be dense in a Hilbert space of functions. We consider and analyze published procedures, both randomized and deterministic, for selecting elements from these families that have been shown to ensure the rate of convergence in $L_2$ norm of order $O(1/N)$, where $N$ is the number of elements. We show that both randomized and deterministic procedures are successful if additional information about the families of functions to be approximated is provided. In the absence of such additional information one may observe exponential growth of the number of terms needed to approximate the function and/or extreme sensitivity of the outcome of the approximation to parameters. Implications of our analysis for applications of neural networks in modeling and control are illustrated with examples., Comment: arXiv admin note: text overlap with arXiv:0905.0677

We discuss the role of random basis function approximators in modeling and control. We analyze the published work on random basis function approximators and demonstrate that their favorable error rate of convergence O(1/n) is guaranteed only with very substantial computational resources. We also discuss implications of our analysis for applications of neural networks in modeling and control.

We address the important theoretical question why a recurrent neural network with fixed weights can adaptively classify time-varied signals in the presence of additive noise and parametric perturbations. We provide a mathematical proof assuming that unknown parameters are allowed to enter the signal nonlinearly and the noise amplitude is sufficiently small., Comment: 22 pages

We propose a technique for the design and analysis of adaptation algorithms in dynamical systems. The technique applies both to systems with conventional Lyapunov-stable target dynamics and to ones of which the desired dynamics around the target set is nonequilibrium and in general unstable in the Lyapunov sense. Mathematical models of uncertainties are allowed to be nonlinearly parametrized, smooth, and monotonic functions of linear functionals of the parameters. We illustrate with applications how the proposed method leads to control algorithms. In particular we show that the mere existence of nonlinear operator gains for the desired dynamics guarantees that system solutions are bounded, reach a neighborhood of the target set, and mismatches between the modeled uncertainties and uncertainty compensator vanish with time. The proposed class of algorithms can also serve as parameter identification procedures. In particular, standard persistent excitation suffices to ensure exponential convergence of the estimated to the actual values of the parameters. When a weak, nonlinear version of the persistent excitation condition is satisfied, convergence is asymptotic. The approach extends to a broader class of parameterizations where the monotonicity restriction holds only locally. In this case excitation with oscillations of sufficiently high frequency ensure convergence., Comment: 40 pages,4 figures, 1 table

We propose novel parameter estimation algorithms for a class of dynamical systems with nonlinear parametrization. The class is initially restricted to smooth monotonic functions with respect to a linear functional of the parameters. We show that under this restriction standard persistent excitation suffices to ensure exponentially fast convergence of the estimates to the actual values of unknown parameters. Subsequently, our approach is extended to cases in which the monotonicity assumption holds only locally. We show that excitation with high-frequency of oscillations is sufficient to ensure convergence. Two practically relevant examples are given in order to illustrate the effectiveness of the approach., Comment: 24 pages, 4 figures

Superposition of sigmoid function over a finite time interval is shown to be equivalent to the linear combination of the solutions of a linearly parameterized system of logistic differential equations. Due to the linearity with respect to the parameters of the system, it is possible to design an effective procedure for parameter adjustment. Stability properties of this procedure are analyzed. Strategies shown in earlier studies to facilitate learning such as randomization of a learning sequence and adding specially designed disturbances during the learning phase are requirements for guaranteeing convergence in the learning scheme proposed., Comment: 30 pages, 7 figures