Search

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.

We compute probabilistic controlled invariant sets for nonlinear systems using Gaussian process state space models, which are data-driven models that account for unmodeled and unknown nonlinear dynamics. We investigate the relationship between robust and probabilistic invariance, leveraging this relationship to design state-feedback controllers that maximize the probability of the system staying within the probabilistic controlled invariant set. We propose a semi-definite-programming-based optimization scheme for designing the state-feedback controllers subject to input constraints. The effectiveness of our results are demonstrated and validated on a quadrotor, both in simulation and on a physical platform.

The automotive industry is transitioning from traditional ECU-based systems to software-defined vehicles. A central role of this revolution is played by containers, lightweight virtualization technologies that enable the flexible consolidation of complex software applications on a common hardware platform. Despite their widespread adoption, the impact of containerization on fundamental real-time metrics such as end-to-end latency, communication jitter, as well as memory and CPU utilization has remained virtually unexplored. This paper presents a microservice architecture for a real-world autonomous driving application where containers isolate each service. Our comprehensive evaluation shows the benefits in terms of end-to-end latency of such a solution even over standard bare-Linux deployments. Specifically, in the case of the presented microservice architecture, the mean end-to-end latency can be improved by 5-8 %. Also, the maximum latencies were significantly reduced using container deployment.

In reinforcement learning (RL), exploiting environmental symmetries can significantly enhance efficiency, robustness, and performance. However, ensuring that the deep RL policy and value networks are respectively equivariant and invariant to exploit these symmetries is a substantial challenge. Related works try to design networks that are equivariant and invariant by construction, limiting them to a very restricted library of components, which in turn hampers the expressiveness of the networks. This paper proposes a method to construct equivariant policies and invariant value functions without specialized neural network components, which we term equivariant ensembles. We further add a regularization term for adding inductive bias during training. In a map-based path planning case study, we show how equivariant ensembles and regularization benefit sample efficiency and performance., Comment: Accepted at IROS 2024. A video can be found here: https://youtu.be/L6NOdvU7n7s. The code is available at https://github.com/theilem/uavSim

The rigid gang task model is based on the idea of executing multiple threads simultaneously on a fixed number of processors to increase efficiency and performance. Although there is extensive literature on global rigid gang scheduling, partitioned approaches have several practical advantages (e.g., task isolation and reduced scheduling overheads). In this paper, we propose a new partitioned scheduling strategy for rigid gang tasks, named strict partitioning. The method creates disjoint partitions of tasks and processors to avoid inter-partition interference. Moreover, it tries to assign tasks with similar volumes (i.e., parallelisms) to the same partition so that the intra-partition interference can be reduced. Within each partition, the tasks can be scheduled using any type of scheduler, which allows the use of a less pessimistic schedulability test. Extensive synthetic experiments and a case study based on Edge TPU benchmarks show that strict partitioning achieves better schedulability performance than state-of-the-art global gang schedulability analyses for both preemptive and non-preemptive rigid gang task sets., Comment: Published in IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2024)

The interactive decision-making in multi-agent autonomous racing offers insights valuable beyond the domain of self-driving cars. Mapless online path planning is particularly of practical appeal but poses a challenge for safely overtaking opponents due to the limited planning horizon. To address this, we introduce RaceMOP, a novel method for mapless online path planning designed for multi-agent racing of F1TENTH cars. Unlike classical planners that rely on predefined racing lines, RaceMOP operates without a map, utilizing only local observations to execute high-speed overtaking maneuvers. Our approach combines an artificial potential field method as a base policy with residual policy learning to enable long-horizon planning. We advance the field by introducing a novel approach for policy fusion with the residual policy directly in probability space. Extensive experiments on twelve simulated racetracks validate that RaceMOP is capable of long-horizon decision-making with robust collision avoidance during overtaking maneuvers. RaceMOP demonstrates superior handling over existing mapless planners and generalizes to unknown racetracks, affirming its potential for broader applications in robotics. Our code is available at http://github.com/raphajaner/racemop., Comment: IEEE/RSJ International Conference on Intelligent Robots and Systems 2024

The F1TENTH autonomous driving platform, consisting of 1:10-scale remote-controlled cars, has evolved into a well-established education and research platform. The many publications and real-world competitions span many domains, from classical path planning to novel learning-based algorithms. Consequently, the field is wide and disjointed, hindering direct comparison of developed methods and making it difficult to assess the state-of-the-art. Therefore, we aim to unify the field by surveying current approaches, describing common methods, and providing benchmark results to facilitate clear comparisons and establish a baseline for future work. This research aims to survey past and current work with F1TENTH vehicles in the classical and learning categories and explain the different solution approaches. We describe particle filter localisation, trajectory optimisation and tracking, model predictive contouring control, follow-the-gap, and end-to-end reinforcement learning. We provide an open-source evaluation of benchmark methods and investigate overlooked factors of control frequency and localisation accuracy for classical methods as well as reward signal and training map for learning methods. The evaluation shows that the optimisation and tracking method achieves the fastest lap times, followed by the online planning approach. Finally, our work identifies and outlines the relevant research aspects to help motivate future work in the F1TENTH domain., Comment: 12 pages, 18 figures. Sumbitted for publication

Cache partitioning techniques have been successfully adopted to mitigate interference among concurrently executing real-time tasks on multi-core processors. Considering that the execution time of a cache-sensitive task strongly depends on the cache available for it to use, co-optimizing cache partitioning and task allocation improves the system's schedulability. In this paper, we propose a hybrid multi-layer design space exploration technique to solve this multi-resource management problem. We explore the interplay between cache partitioning and schedulability by systematically interleaving three optimization layers, viz., (i) in the outer layer, we perform a breadth-first search combined with proactive pruning for cache partitioning; (ii) in the middle layer, we exploit a first-fit heuristic for allocating tasks to cores; and (iii) in the inner layer, we use the well-known recurrence relation for the schedulability analysis of non-preemptive fixed-priority (NP-FP) tasks in a uniprocessor setting. Although our focus is on NP-FP scheduling, we evaluate the flexibility of our framework in supporting different scheduling policies (NP-EDF, P-EDF) by plugging in appropriate analysis methods in the inner layer. Experiments show that, compared to the state-of-the-art techniques, the proposed framework can improve the real-time schedulability of NP-FP task sets by an average of 15.2% with a maximum improvement of 233.6% (when tasks are highly cache-sensitive) and a minimum of 1.6% (when cache sensitivity is low). For such task sets, we found that clustering similar-period (or mutually compatible) tasks often leads to higher schedulability (on average 7.6%) than clustering by cache sensitivity. In our evaluation, the framework also achieves good results for preemptive and dynamic-priority scheduling policies., Comment: to be published in IEEE Real-Time Systems Symposium (RTSS), 2023

Coverage path planning (CPP) is a critical problem in robotics, where the goal is to find an efficient path that covers every point in an area of interest. This work addresses the power-constrained CPP problem with recharge for battery-limited unmanned aerial vehicles (UAVs). In this problem, a notable challenge emerges from integrating recharge journeys into the overall coverage strategy, highlighting the intricate task of making strategic, long-term decisions. We propose a novel proximal policy optimization (PPO)-based deep reinforcement learning (DRL) approach with map-based observations, utilizing action masking and discount factor scheduling to optimize coverage trajectories over the entire mission horizon. We further provide the agent with a position history to handle emergent state loops caused by the recharge capability. Our approach outperforms a baseline heuristic, generalizes to different target zones and maps, with limited generalization to unseen maps. We offer valuable insights into DRL algorithm design for long-horizon problems and provide a publicly available software framework for the CPP problem., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

Directed acyclic graph (DAG) tasks are currently adopted in the real-time domain to model complex applications from the automotive, avionics, and industrial domains that implement their functionalities through chains of intercommunicating tasks. This paper studies the problem of scheduling real-time DAG tasks by presenting a novel schedulability test based on the concept of trivial schedulability. Using this schedulability test, we propose a new DAG scheduling framework (edge generation scheduling -- EGS) that attempts to minimize the DAG width by iteratively generating edges while guaranteeing the deadline constraint. We study how to efficiently solve the problem of generating edges by developing a deep reinforcement learning algorithm combined with a graph representation neural network to learn an efficient edge generation policy for EGS. We evaluate the effectiveness of the proposed algorithm by comparing it with state-of-the-art DAG scheduling heuristics and an optimal mixed-integer linear programming baseline. Experimental results show that the proposed algorithm outperforms the state-of-the-art by requiring fewer processors to schedule the same DAG tasks. The code is available at https://github.com/binqi-sun/egs., Comment: Accepted for publication in IEEE Transactions on Computers

In this paper, we present the synthesis of secure-by-construction controllers that address safety and security properties simultaneously in cyber-physical systems. Our focus is on studying a specific security property called opacity, which characterizes the system's ability to maintain plausible deniability of its secret behavior in the presence of an intruder. These controllers are synthesized based on a concept of so-called (augmented) control barrier functions, which we introduce and discuss in detail. We propose conditions that facilitate the construction of the desired (augmented) control barrier functions and their corresponding secure-by-construction controllers. To compute these functions, we propose an iterative scheme that leverages iterative sum-of-square programming techniques. This approach enables efficient computation of these functions, particularly for polynomial systems. Moreover, we demonstrate the flexibility of our approach by incorporating user-defined cost functions into the construction of secure-by-construction controllers. Finally, we validate the effectiveness of our results through two case studies, illustrating the practical applicability and benefits of our proposed approach.

Deploying teams of unmanned aerial vehicles (UAVs) to harvest data from distributed Internet of Things (IoT) devices requires efficient trajectory planning and coordination algorithms. Multi-agent reinforcement learning (MARL) has emerged as a solution, but requires extensive and costly real-world training data. To tackle this challenge, we propose a novel model-aided federated MARL algorithm to coordinate multiple UAVs on a data harvesting mission with only limited knowledge about the environment. The proposed algorithm alternates between building an environment simulation model from real-world measurements, specifically learning the radio channel characteristics and estimating unknown IoT device positions, and federated QMIX training in the simulated environment. Each UAV agent trains a local QMIX model in its simulated environment and continuously consolidates it through federated learning with other agents, accelerating the learning process. A performance comparison with standard MARL algorithms demonstrates that our proposed model-aided FedQMIX algorithm reduces the need for real-world training experiences by around three magnitudes while attaining similar data collection performance.

Autonomous driving involves complex decision-making in highly interactive environments, requiring thoughtful negotiation with other traffic participants. While reinforcement learning provides a way to learn such interaction behavior, efficient learning critically depends on scalable state representations. Contrary to imitation learning methods, high-dimensional state representations still constitute a major bottleneck for deep reinforcement learning methods in autonomous driving. In this paper, we study the challenges of constructing bird's-eye-view representations for autonomous driving and propose a recurrent learning architecture for long-horizon driving. Our PPO-based approach, called RecurrDriveNet, is demonstrated on a simulated autonomous driving task in CARLA, where it outperforms traditional frame-stacking methods while only requiring one million experiences for efficient training. RecurrDriveNet causes less than one infraction per driven kilometer by interacting safely with other road users., Comment: IEEE International Conference on Intelligent Transportation Systems 2023

This paper proposes the Phy-DRL: a physics-regulated deep reinforcement learning (DRL) framework for safety-critical autonomous systems. The Phy-DRL has three distinguished invariant-embedding designs: i) residual action policy (i.e., integrating data-driven-DRL action policy and physics-model-based action policy), ii) automatically constructed safety-embedded reward, and iii) physics-model-guided neural network (NN) editing, including link editing and activation editing. Theoretically, the Phy-DRL exhibits 1) a mathematically provable safety guarantee and 2) strict compliance of critic and actor networks with physics knowledge about the action-value function and action policy. Finally, we evaluate the Phy-DRL on a cart-pole system and a quadruped robot. The experiments validate our theoretical results and demonstrate that Phy-DRL features guaranteed safety compared to purely data-driven DRL and solely model-based design while offering remarkably fewer learning parameters and fast training towards safety guarantee.

Deep reinforcement learning (DRL) has achieved tremendous success in many complex decision-making tasks of autonomous systems with high-dimensional state and/or action spaces. However, the safety and stability still remain major concerns that hinder the applications of DRL to safety-critical autonomous systems. To address the concerns, we proposed the Phy-DRL: a physical deep reinforcement learning framework. The Phy-DRL is novel in two architectural designs: i) Lyapunov-like reward, and ii) residual control (i.e., integration of physics-model-based control and data-driven control). The concurrent physical reward and residual control empower the Phy-DRL the (mathematically) provable safety and stability guarantees. Through experiments on the inverted pendulum, we show that the Phy-DRL features guaranteed safety and stability and enhanced robustness, while offering remarkably accelerated training and enlarged reward., Comment: Working Paper

Gear assembly is an essential but challenging task in industrial automation. This paper presents a novel two-stage approach for achieving high-precision and flexible gear assembly. The proposed approach integrates YOLO to coarsely localize the workpiece in a searching phase and deep reinforcement learning (DRL) to complete the insertion. Specifically, DRL addresses the challenge of partial visibility when the on-wrist camera is too close to the workpiece. Additionally, force feedback is used to smoothly transit the process from the first phase to the second phase. To reduce the data collection effort for training deep neural networks, we use synthetic RGB images for training YOLO and construct an offline interaction environment leveraging sampled real-world data for training DRL agents. We evaluate the proposed approach in a gear assembly experiment with a precision tolerance of 0.3mm. The results show that our method can robustly and efficiently complete searching and insertion from arbitrary positions within an average of 15 seconds., Comment: Submitted to 2023 IEEE/RSJ International Conference on Intelligent Robots (IROS)

The development of vehicle controllers for autonomous racing is challenging because racing cars operate at their physical driving limit. Prompted by the demand for improved performance, autonomous racing research has seen the proliferation of machine learning-based controllers. While these approaches show competitive performance, their practical applicability is often limited. Residual policy learning promises to mitigate this drawback by combining classical controllers with learned residual controllers. The critical advantage of residual controllers is their high adaptability parallel to the classical controller's stable behavior. We propose a residual vehicle controller for autonomous racing cars that learns to amend a classical controller for the path-following of racing lines. In an extensive study, performance gains of our approach are evaluated for a simulated car of the F1TENTH autonomous racing series. The evaluation for twelve replicated real-world racetracks shows that the residual controller reduces lap times by an average of 4.55 % compared to a classical controller and even enables lap time gains on unknown racetracks., Comment: IEEE Intelligent Vehicles Symposium 2023

Modern cyber-physical systems are becoming increasingly complex to model, thus motivating data-driven techniques such as reinforcement learning (RL) to find appropriate control agents. However, most systems are subject to hard constraints such as safety or operational bounds. Typically, to learn to satisfy these constraints, the agent must violate them systematically, which is computationally prohibitive in most systems. Recent efforts aim to utilize feasibility models that assess whether a proposed action is feasible to avoid applying the agent's infeasible action proposals to the system. However, these efforts focus on guaranteeing constraint satisfaction rather than the agent's learning efficiency. To improve the learning process, we introduce action mapping, a novel approach that divides the learning process into two steps: first learn feasibility and subsequently, the objective by mapping actions into the sets of feasible actions. This paper focuses on the feasibility part by learning to generate all feasible actions through self-supervised querying of the feasibility model. We train the agent by formulating the problem as a distribution matching problem and deriving gradient estimators for different divergences. Through an illustrative example, a robotic path planning scenario, and a robotic grasping simulation, we demonstrate the agent's proficiency in generating actions across disconnected feasible action sets. By addressing the feasibility step, this paper makes it possible to focus future work on the objective part of action mapping, paving the way for an RL framework that is both safe and efficient.

Advances in deep learning have revolutionized cyber-physical applications, including the development of Autonomous Vehicles. However, real-world collisions involving autonomous control of vehicles have raised significant safety concerns regarding the use of Deep Neural Networks (DNN) in safety-critical tasks, particularly Perception. The inherent unverifiability of DNNs poses a key challenge in ensuring their safe and reliable operation. In this work, we propose Perception Simplex (PS), a fault-tolerant application architecture designed for obstacle detection and collision avoidance. We analyze an existing LiDAR-based classical obstacle detection algorithm to establish strict bounds on its capabilities and limitations. Such analysis and verification have not been possible for deep learning-based perception systems yet. By employing verifiable obstacle detection algorithms, PS identifies obstacle existence detection faults in the output of unverifiable DNN-based object detectors. When faults with potential collision risks are detected, appropriate corrective actions are initiated. Through extensive analysis and software-in-the-loop simulations, we demonstrate that PS provides predictable and deterministic fault tolerance against obstacle existence detection faults, establishing a robust safety guarantee., Comment: arXiv admin note: substantial text overlap with arXiv:2208.14403

6D pose recognition has been a crucial factor in the success of robotic grasping, and recent deep learning based approaches have achieved remarkable results on benchmarks. However, their generalization capabilities in real-world applications remain unclear. To overcome this gap, we introduce 6IMPOSE, a novel framework for sim-to-real data generation and 6D pose estimation. 6IMPOSE consists of four modules: First, a data generation pipeline that employs the 3D software suite Blender to create synthetic RGBD image datasets with 6D pose annotations. Second, an annotated RGBD dataset of five household objects generated using the proposed pipeline. Third, a real-time two-stage 6D pose estimation approach that integrates the object detector YOLO-V4 and a streamlined, real-time version of the 6D pose estimation algorithm PVN3D optimized for time-sensitive robotics applications. Fourth, a codebase designed to facilitate the integration of the vision system into a robotic grasping experiment. Our approach demonstrates the efficient generation of large amounts of photo-realistic RGBD images and the successful transfer of the trained inference model to robotic grasping experiments, achieving an overall success rate of 87% in grasping five different household objects from cluttered backgrounds under varying lighting conditions. This is made possible by the fine-tuning of data generation and domain randomization techniques, and the optimization of the inference pipeline, overcoming the generalization and performance shortcomings of the original PVN3D algorithm. Finally, we make the code, synthetic dataset, and all the pretrained models available on Github.

Perception of obstacles remains a critical safety concern for autonomous vehicles. Real-world collisions have shown that the autonomy faults leading to fatal collisions originate from obstacle existence detection. Open source autonomous driving implementations show a perception pipeline with complex interdependent Deep Neural Networks. These networks are not fully verifiable, making them unsuitable for safety-critical tasks. In this work, we present a safety verification of an existing LiDAR based classical obstacle detection algorithm. We establish strict bounds on the capabilities of this obstacle detection algorithm. Given safety standards, such bounds allow for determining LiDAR sensor properties that would reliably satisfy the standards. Such analysis has as yet been unattainable for neural network based perception systems. We provide a rigorous analysis of the obstacle detection system with empirical results based on real-world sensor data., Comment: Accepted at ISSRE 2022

In this paper, we provide a direct data-driven approach to synthesize safety controllers for unknown linear systems affected by unknown-but-bounded disturbances, in which identifying the unknown model is not required. First, we propose a notion of $\gamma$-robust safety invariant ($\gamma$-RSI) sets and their associated state-feedback controllers, which can be applied to enforce invariance properties. Then, we formulate a data-driven computation of these sets in terms of convex optimization problems with linear matrix inequalities (LMI) as constraints, which can be solved based on a finite number of data collected from a single input-state trajectory of the system. To show the effectiveness of the proposed approach, we apply our results to a 4-dimensional inverted pendulum., Comment: 6th IEEE Conference on Control Technology and Applications

In this paper, we propose a construction scheme for a Safe-visor architecture for sandboxing unverified controllers, e.g., artificial intelligence-based (a.k.a. AI-based) controllers, in two-players non-cooperative stochastic games. Concretely, we leverage abstraction-based approaches to construct a supervisor that checks and decides whether or not to accept the inputs provided by the unverified controller, and a safety advisor that provides fallback control inputs to ensure safety whenever the unverified controller is rejected. Moreover, by leveraging an ($\epsilon,\delta$)-approximate probabilistic relation between the original game and its finite abstraction, we provide a formal safety guarantee with respect to safety specifications modeled by deterministic finite automata (DFA), while the functionality of the unverified controllers is still exploited. To show the effectiveness of the proposed results, we apply them to a control problem of a quadrotor tracking a moving ground vehicle, in which an AI-based unverified controller is employed to control the quadrotor.

Deep reinforcement learning (DRL) is a promising approach to solve complex control tasks by learning policies through interactions with the environment. However, the training of DRL policies requires large amounts of training experiences, making it impractical to learn the policy directly on physical systems. Sim-to-real approaches leverage simulations to pretrain DRL policies and then deploy them in the real world. Unfortunately, the direct real-world deployment of pretrained policies usually suffers from performance deterioration due to the different dynamics, known as the reality gap. Recent sim-to-real methods, such as domain randomization and domain adaptation, focus on improving the robustness of the pretrained agents. Nevertheless, the simulation-trained policies often need to be tuned with real-world data to reach optimal performance, which is challenging due to the high cost of real-world samples. This work proposes a distributed cloud-edge architecture to train DRL agents in the real world in real-time. In the architecture, the inference and training are assigned to the edge and cloud, separating the real-time control loop from the computationally expensive training loop. To overcome the reality gap, our architecture exploits sim-to-real transfer strategies to continue the training of simulation-pretrained agents on a physical system. We demonstrate its applicability on a physical inverted-pendulum control system, analyzing critical parameters. The real-world experiments show that our architecture can adapt the pretrained DRL agents to unseen dynamics consistently and efficiently., Comment: Submitted to IROS 2022

In this paper, we present how to synthesize controllers to enforce $\omega$-regular properties over linear control systems affected by bounded disturbances. In particular, these controllers are synthesized based on so-called hybrid controlled invariant (HCI) sets. To compute these sets, we first construct a product system between the linear control system and the deterministic Streett automata (DSA) modeling the desired property. Then, we compute the maximal HCI set over the state set of the product system by leveraging a set-based approach. To ensure termination of the computation of the HCI sets within a finite number of iterations, we also propose two iterative schemes to compute approximations of the maximal HCI set. Finally, we show the effectiveness of our results via two case studies.