IROS 2024 论文列表

As transportation systems see gradual deployment of connected and automated vehicles (CAVs), there is increasing opportunity for intelligent coordination of CAVs towards system-wide objectives. While numerous previous works have modeled single junctions (e.g. intersections and merges) and investigated control theory-based strategies for vehicle-based coordination, this work investigates the scalability of vehicular control approaches to large networks of intersections, where interactions among multiple intersections may amplify traffic disturbances. Moreover, this work focuses on mixed autonomy networks where the highly nonlinear behavior of human-driven vehicles (HDVs) complicate overall system dynamics, and where the formation of CAV-led platoons may be advantageous. Two approaches are considered for the studied settings: model predictive control (MPC) and model-free reinforcement learning (RL), both adapted from previous methods designed for single intersection and/or full autonomy settings. Results in a network of two intersections demonstrate that MPC faces significant challenges in low-level nonlinear trajectory optimization as well as high-level crossing scheduling, while the RL policies implicitly optimizes for both low-level control and high-level coordination. Scalability analysis in large networks with hundred of intersections reveal that policies derived from additional finetuning only suffer mild degradation in performance despite the numerous out-of-distribution traffic conditions that may emerge under large scale.

This paper showcases the application of the Saturation in the Null Space (SNS) algorithm to establish task prioritization and coordination within a tele-operated minimally invasive robotic surgical setting. In our work, SNS prioritizes achieving Remote Center of Motion (RCM) constraint, ensuring safe instrument manipulation, while respecting joint constraints for uninterrupted robot operation. This prioritization allows for accommodating the tracking of the surgeon’s motion, within the capabilities defined by RCM and joint constraints. We investigate both the velocity and acceleration control variants of the SNS algorithm, incorporating bespoke adjustments to tailor the original algorithm to the intricate requirements of surgical applications. Through simulations and experiments, this work aims to demonstrate the effectiveness of SNS in enhancing the safety and controllability of tele-operated surgery, paving the way for its integration in various surgical procedures.

One core component of Active Debris Removal (ADR) and On-Orbit Servicing (OOS) missions in space is to estimate and track the relative pose of a non-cooperative satellite in close proximity. Conventionally, Image Processing methods have been popular in pose estimation by employing manual feature extraction techniques. But the performance of such methods plateaus in the challenging illumination conditions and sensor capability constraints in space, because of which Deep Learning (DL)-based approaches have gained traction. This paper aims to provide an improvement over the existing state-of-the-art direct pose estimation methods from a monocular camera, without relying on any 3D model of the target satellite. The main contribution of this work is to develop a general purpose satellite-invariant pose estimation architecture with improved accuracy and implement an adaptive navigation filter over it to track the pose continuously over a stream of images. The pose estimation module includes a modified DenseNet architecture. In order to test the generalization capability, the proposed pose estimation module is tested on the SPEED, SPEED+, SHIRT and URSO datasets and compared with other existing methods. The advantage of the proposed method is that the same model architecture is able to give accurate pose estimation results for different satellite datasets. To perform continuous tracking of the relative pose, an adaptive EKF (Extended Kalman Filter) is implemented on the initial pose estimates. For performance evaluation of the navigation filter, the accuracy goals required for the relative navigation of Hubble Space Telescope SM4 mission are considered while testing on the SHIRT dataset.

Motion planning is a fundamental capability in robotics applications. Real-world scenarios can introduce uncertainty to the motion planning problem. In this work we study environment uncertainty in general high-dimensional problems wherein the choice of appropriate metrics and formulations are shown to have significant effect on the probability of collision of the solution path. Several practically motivated cost functions have been proposed in literature to model and solve the problem but are shown in this work to suffer from higher probabilities of collision. The current work presents a theoretically sound formulation that was first mentioned in previous work on minimum constraint removal. In this work, approximating the optimal problem is shown to be better in achieving lower probability of collision. To demonstrate the formulation in a sampling-based setting, a mixed integer linear program seeded by greedy search over a roadmap with sampled environments is used to report paths with low probability of collision. Compared against minimizing the sum and minimizing max probability cost functions on a seven degree-of-freedom robotic arm in uncertain environments, we show clear benefits and promise towards motion planning for optimal probability of collision.

This paper proposes a comprehensive framework designed for the autonomous inspection of complex environments, with a specific focus on multi-tiered settings such as distillation column trays. Leveraging quadruped robots equipped with roller arms, and through the use of onboard perception, we integrate essential motion components including: locomotion, safe and dynamic transitions between trays, and intermediate motions that bridge a variety of motion primitives. Given the slippery and confined nature of column trays, it is critical to ensure safety of the robot during inspection, therefore we employ a safety filter and footstep re-planning based upon control barrier function representations of the environment. Our framework integrates all system components into a state machine encoding the developed safety-critical planning and control elements to guarantee safety-critical autonomy, enabling autonomous and safe navigation and inspection of distillation columns. Experimental validation in an environment, consisting of industrial-grade chemical distillation trays, highlights the effectiveness of our multi-layered architecture.

This paper introduces a Safety-First Tracker (SF-Tracker) designed for omnidirectional autonomous tracking robots. The position and orientation of omnidirectional robots are decoupled for stepwise planning to ensure trajectory safety and maintain target visibility. SF-Tracker puts the trajectory safety in the first place. First, a collision-free and occlusion-free reference path is efficiently initialized by constructing a directed weighted graph. By building upon this path, safe trajectory optimization is implemented to ensure safe movement. Finally, an orientation planner is developed to achieve target visibility based on the safe trajectory. Extensive experimental evaluations in simulated environments and the real world demonstrate that the SF-Tracker outperforms state-of-the-art methods in terms trajectory safety and target visibility. Ablation experiments further demonstrate the significance of each step of the SF-Tracker. The source code and demonstration video can be found at https://github.com/Yue-0/SF-Tracker.

Bimanual dexterous manipulation in robotics, essential for a wide range of applications, addresses the critical challenge of balancing intricate operational capabilities with assured safety and reliability. While Safe Reinforcement Learning is integral to the robustness of robotic systems, the area of safe multi-agent reinforcement learning (MARL), cooperative control of multiple robots has been scarcely studied. In this study, we explore MARL for safe cooperative control with multiple robot hands. Each robot must follow individual and collective safety guidelines to ensure safe team actions. However, the non-stationarity inherent in current algorithms hinders the precise updating of strategies to satisfy these safety constraints effectively. In this paper, we propose Multi-Agent Constrained Proximal Advantage Optimization (MAC-PAO), which considers the sequence of agent updates and integrates non-stationarity into sequential update schemes. This algorithm ensures consistent improvement in both rewards and adherence to safety constraints in each iteration. We tested MACPAO on various tasks with safety constraints and demonstrated that it outperforms other MARL algorithms in balancing reward enhancement and safety compliance. Supplementary materials and code are available at the provided link https://github.com/YONEX4090/MultiSafeHand.git.

Recent advances in both legged robot locomotion and Reinforcement Learning have shown a promising path for developing bipedal robot controllers. While the difference in dynamics between real world and simulation, also known as reality gap, still hinders the use. In this paper, we focus on sim-to-real bipedal robot locomotion task. We leverage the recent advances in auto-tuning sim-to-real transfer and use it to address sim-to-real bipedal robot locomotion problem. Similar to existing work, we first train a parameter searching model with dataset collected from simulator and use real-world data to tune the simulation parameters. However, the prediction tuning can be unreliable if the training dataset distribution fails to cover the real-world data. We address this problem by formulating this problem as an Out-of-distribution problem and further extending the current framework with a dataset verification model. With extended module, our method is capable of tuning the simulation parameters safely and efficiently. We demonstrate our method outperforms existing work and achieves sim-to-real bipedal robot locomotion on bipedal robot BITeno.

Ensuring safety in Reinforcement Learning (RL), typically framed as a Constrained Markov Decision Process (CMDP), is crucial for real-world exploration applications. Current approaches in handling CMDP struggle to balance optimality and feasibility, as direct optimization methods can-not ensure state-wise in-training safety, and projection-based methods correct actions inefficiently through lengthy iterations. To address these challenges, we propose Adaptive Chance-constrained Safeguards (ACS), an adaptive, model-free safe RL algorithm using the safety recovery rate as a surrogate chance constraint to iteratively ensure safety during exploration and after achieving convergence. Theoretical analysis indicates that the relaxed probabilistic constraint sufficiently guarantees forward invariance to the safe set. And extensive experiments conducted on both simulated and real-world safety-critical tasks demonstrate its effectiveness in enforcing safety (nearly zero-violation) while preserving optimality (+23.8%), robustness, and fast response in stochastic real-world settings.

We introduce the Safe Offline-to-Online Multi-Agent Decision Transformer (SO2-MADT), an innovative framework that revolutionizes safety considerations in Multi-agent Reinforcement Learning (MARL) through a novel sequence modeling approach. Leveraging the dynamic capabilities inherent in Decision Transformers, our methodology seamlessly incorporates safety protocols as a cornerstone element, ensuring secure operations throughout both the offline pre-training phase and the adaptive online fine-tuning phase. At the core of our framework lie two pivotal innovations: the Safety-To-Go (STG) token, embedding safety at a macro level, and the Agent Prioritization Module (APM), facilitating explicit credit assignment at a micro level. Through extensive testing against the challenging environments of the StarCraft Multi-Agent Challenge (SMAC) and Multi-agent MuJoCo, our SO2-MADT not only excels in offline pre-training but also demonstrates superior performance during online fine-tuning, without any degradation in performance. The implications of our work provide a pathway for deployment in critical real-world applications where safety is paramount and non-negotiable. The code is available at https://github.com/shahaamirbader/SO2-MADT.

Flexible robots may overcome some of the industry’s major challenges, such as enabling intrinsically safe human-robot collaboration and achieving a higher payload-to-mass ratio. However, controlling flexible robots is complicated due to their complex dynamics, which include oscillatory behavior and a high-dimensional state space. Nonlinear model predictive control (NMPC) offers an effective means to control such robots, but its significant computational demand often limits its application in real-time scenarios. To enable fast control of flexible robots, we propose a framework for a safe approximation of NMPC using imitation learning and a predictive safety filter. Our framework significantly reduces computation time while incurring a slight loss in performance. Compared to NMPC, our framework shows more than an eightfold improvement in computation time when controlling a three-dimensional flexible robot arm in simulation, all while guaranteeing safety constraints. Notably, our approach out-performs state-of-the-art reinforcement learning methods. The development of fast and safe approximate NMPC holds the potential to accelerate the adoption of flexible robots in industry. The project code is available at: tinyurl.com/anmpc4fr

In the realm of autonomous agents, ensuring safety and reliability in complex and dynamic environments remains a paramount challenge. Safe reinforcement learning addresses these concerns by introducing safety constraints, but still faces challenges in navigating intricate environments such as complex driving situations. To overcome these challenges, we present the safe constraint reward (Safe CoR) framework, a novel method that utilizes two types of expert demonstrations—reward expert demonstrations focusing on performance optimization and safe expert demonstrations prioritizing safety. By exploiting a constraint reward (CoR), our framework guides the agent to balance performance goals of reward sum with safety constraints. We test the proposed framework in diverse environments, including the safety gym, metadrive, and the real-world Jackal platform. Our proposed framework improves algorithm performance by 39% and reduces constraint violations by 88% on the real-world Jackal platform, highlighting its effectiveness. Through this innovative approach, we expect significant advancements in real-world performance, leading to transformative effects in the realm of safe and reliable autonomous agents.

Predicting agents’ trajectories in complex environments is critical for achieving safe autonomous robot navigation. Empirically, agents’ decisions and preferences are susceptible to changes in environmental factors (e.g., interactions with other agents, weather conditions, traffic rules). State-of-the-art methods rely on High-Definition (HD) or semantic maps to model the environment, but do not take into account unpredictable factors such as complex weather conditions. In addition, since HD maps are nontrivial to obtain, those methods are limited in the scope of environments they can be applied in. We propose a more flexible graph based trajectory prediction model that uses only images to model the environment, without requiring expensive map information. We experimentally validate our proposed model, demonstrating robust performances in trajectory prediction compared to state-of-the-art methods, and outperform in complex environments that cannot be modeled with purely map based methods, such as diverse weather conditions.

Large-scale point cloud consists of a multitude of individual objects, thereby encompassing rich structural and underlying semantic contextual information, resulting in a challenging problem in efficiently segmenting a point cloud. Most existing researches mainly focus on capturing intricate local features without giving due consideration to global ones, thus failing to leverage semantic context. In this paper, we propose a Similarity-Weighted Convolution and local-global Fusion Network, named SWCF-Net, which takes into account both local and global features. We propose a Similarity-Weighted Convolution (SWConv) to effectively extract local features, where similarity weights are incorporated into the convolution operation to enhance the generalization capabilities. Then, we employ a downsampling operation on the K and V channels within the attention module, thereby reducing the quadratic complexity to linear, enabling Transformer to deal with large-scale point cloud. At last, orthogonal components are extracted in the global features and then aggregated with local features, thereby eliminating redundant information between local and global features and consequently promoting efficiency. We evaluate SWCF-Net on large-scale outdoor datasets SemanticKITTI and Toronto3D. Our experimental results demonstrate the effectiveness of the proposed network. Our method achieves a competitive result with less computational cost, and is able to handle large-scale point clouds efficiently. The code is available at https://github.com/Sylva-Lin/SWCF-Net.

Inaccurate tool localization is one of the main reasons for failures in automating surgical tasks. Imprecise robot kinematics and noisy observations caused by the poor visual acuity of an endoscopic camera make tool tracking challenging. Previous works in surgical automation adopt environment-specific setups or hard-coded strategies instead of explicitly considering motion and observation uncertainty of tool tracking in their policies. In this work, we present SURESTEP, an uncertainty-aware trajectory optimization framework for robust surgical automation.We model the uncertainty in tool tracking by considering noise sources that are typical in surgical environments.Using a Gaussian assumption to propagate our uncertainty models through a given tool trajectory, SURESTEP provides a general framework that minimizes the upper bound on the entropy of the final estimated tool distribution.We showcase our method by performing the first-ever, to our knowledge, needle regrasping with a moving endoscopic camera.We compare SURESTEP with a baseline method on a real-world suture needle regrasping task under challenging environmental conditions, such as poor lighting and a moving endoscopic camera. The results over 60 regrasps on the da Vinci Research Kit (dVRK) demonstrate that our optimized trajectories significantly outperform the un-optimized baseline.

Most RGB-D-based SLAM methods assume texture-rich environments, making them susceptible to significant tracking errors or complete failures in the absence of texture features. Moreover, many existing methods encounter substantial rotation estimation errors, leading to long-term drift in tracking. This paper proposes a novel structured-constrained RGB-D SLAM method (STL-SLAM) for texture-limited environments. Compared to the existing methods, STL-SLAM can deal with environments without abundant texture information and significantly reduce long-term drift caused by rotation estimation errors. We assess the distribution complexity of pixels in an image by calculating the information entropy and pre-processing accordingly. We also present an efficient Manhattan Frames (MF) detection strategy based on orthogonal planes and lines. If MF is detected, we decouple rotation and translation, estimate drift-free rotation based on the Manhattan World (MW) coordinate system, and then estimate translation by minimizing the re-projection error of point, line, and plane features. In non-Manhattan Frames, the 6-DoF pose estimation is performed holistically, with the incorporation of structural constraints of parallel and perpendicular planes, as well as parallel and vertical lines, into the optimization process. Finally, we evaluate our method on public datasets and in real-world environments, which shows that our proposed method achieves superior performance compared to its counterparts.

Many autonomous robotic applications require object-level understanding when deployed. Actively reconstructing objects of interest, i.e. objects with specific semantic meanings, is therefore relevant for a robot to perform downstream tasks in an initially unknown environment. In this work, we propose a novel framework for semantic-targeted active reconstruction using posed RGB-D measurements and 2D semantic labels as input. The key components of our framework are a semantic implicit neural representation and a compatible planning utility function based on semantic rendering and uncertainty estimation, enabling adaptive view planning to target objects of interest. Our planning approach achieves better reconstruction performance in terms of mesh and novel view rendering quality compared to implicit reconstruction baselines that do not consider semantics for view planning. Our framework further outperforms a state-of-the-art semantic-targeted active reconstruction pipeline based on explicit maps, justifying our choice of utilising implicit neural representations to tackle semantic-targeted active reconstruction problems.

The scarcity of annotated infrared (IR) image datasets limits deep learning networks from achieving performances comparable to those achieved with RGB data. To address this, we introduce a novel semi-supervised RGB-to-IR Image-to-Image Translation model (SSL-RGB2IR) that generates synthetic IR data from RGB images. Our model effectively preserves the IR characteristics in the generated images from both synthetic and real-world data. Compared to existing image-to-image translation techniques, training models on this generated IR data significantly improves performance in downstream tasks like segmentation and detection. Notably, in sim-to-real transfer, the segmentation model trained on SSL-RGB2IR generated IR images outperforms baselines and other Image-to-Image (I2I) models. Furthermore, for real-world applications utilizing EO/IR fusion images, this approach solves the well-known challenge of co-registering EO and IR images, which often have inherent misalignment’s due to differing sensor characteristics. Our code is available at https://github.com/prahlad-anand/ssl-rgb2ir https://github.com/prahlad-anand/ssl-rgb2ir.

Monocular depth estimation (MDE) has advanced significantly, primarily through the integration of convolutional neural networks (CNNs) and more recently, Transformers. However, concerns about their susceptibility to adversarial attacks have emerged, especially in safety-critical domains like autonomous driving and robotic navigation. Existing approaches for assessing CNN-based depth prediction methods have fallen short in inducing comprehensive disruptions to the vision system, often limited to specific local areas. In this paper, we introduce SSAP (Shape-Sensitive Adversarial Patch), a novel approach designed to comprehensively disrupt monocular depth estimation (MDE) in autonomous navigation applications. Our patch is crafted to selectively undermine MDE in two distinct ways: by distorting estimated distances or by creating the illusion of an object disappearing from the system’s perspective. Notably, our patch is shape-sensitive, meaning it considers the specific shape and scale of the target object, thereby extending its influence beyond immediate proximity. Furthermore, our patch is trained to effectively address different scales and distances from the camera. Experimental results demonstrate that our approach induces a mean depth estimation error surpassing 0.5, impacting up to 99% of the targeted region for CNN-based MDE models. Additionally, we investigate the vulnerability of Transformer-based MDE models to patch-based attacks, revealing that SSAP yields a significant error of 0.59 and exerts substantial influence over 99% of the target region on these models.