ICRA 2024 论文列表

While individual robots are becoming increasingly capable, the complexity of expected missions increases exponentially in comparison. To cope with this complexity, heterogeneous teams of robots have become a significant research interest in recent years. Making effective use of the robots and their unique skills in a team is challenging. Dynamic runtime conditions often make static task allocations infeasible, requiring a dynamic, capability-aware allocation of tasks to team members. To this end, we propose and implement a system that allows a user to specify missions using Behavior Trees (BTs), which can then, at runtime, be dynamically allocated to the current robot team. The system allows to statically model an individual robot’s capabilities within our ros_bt_py BT framework. It offers a runtime auction system to dynamically allocate tasks to the most capable robot in the current team. The system leverages utility values and pre-conditions to ensure that the allocation improves the overall mission execution quality while preventing faulty assignments. To evaluate the system, we simulated a find-and-decontaminate mission with a team of three heterogeneous robots and analyzed the utilization and overall mission times as metrics. Our results show that our system can improve the overall effectiveness of a team while allowing for intuitive mission specification and flexibility in the team composition.

Robotic systems for manipulation tasks are increasingly expected to be easy to configure for new tasks. While in the past, robot programs were often written statically and tuned manually, the current, faster transition times call for robust, modular and interpretable solutions that also allow a robotic system to learn how to perform a task. We propose the method Behavior-based Bayesian Optimization and Planning (BeBOP) that combines two approaches for generating behavior trees: we build the structure using a reactive planner and learn specific parameters with Bayesian optimization. The method is evaluated on a set of robotic manipulation benchmarks and is shown to outperform state-of-the-art reinforcement learning algorithms by being up to 46 times faster while simultaneously being less dependent on reward shaping. We also propose a modification to the uncertainty estimate for the random forest surrogate models that drastically improves the results.

With the increasing demand for multi-robot exploration of unknown environments, how to accomplish this problem efficiently has become a focus of research. However, in this kind of task, the formulation of strategies for frontier point detection and task allocation largely determines the overall efficiency of the system. In the task of multi-robot exploration of unknown environments, the strategies of frontier point detection and task assignment determine the overall efficiency of the system. Most of the existing methods implement frontier point detection based on the Rapidly-Exploring Random Tree (RRT) and use greedy algorithms for task allocation. However, the classical RRT algorithm is a fixed growth step, which leads to the difficulty of growing branches in narrow environments, making the efficiency and correctness of detecting frontier points lower. Meanwhile, the allocation strategy of the greedy algorithm causes each robot to consider only the exploration area with the largest gain for itself, which easily leads to repeated exploration and reduces the overall efficiency of the system. To solve these problems, we propose an adaptive RRT tree growth strategy for frontier point detection, which can adjust the step size according to the known map information and thus improve the efficiency and accuracy of detection; and introduce a Bayesian-guided evolutionary strategy(BGE) for efficient task allocation, which can utilize the current and historical information to find the optimal allocation scheme in a global perspective. We conduct a comprehensive test of the proposed strategy in the ROS system as well as in the real world, which proves the efficiency of our strategy. Our code is open-sourced and can be provided under request.

It is crucial for robots to be aware of the presence of constraints in order to acquire safe policies. However, explicitly specifying all constraints in an environment can be a challenging task. State-of-the-art constraint inference algorithms learn constraints from demonstrations, but tend to be computationally expensive and prone to instability issues. In this paper, we propose a novel Bayesian method that infers constraints based on preferences over demonstrations. The main advantages of our proposed approach are that it 1) infers constraints without calculating a new policy at each iteration, 2) uses a simple and more realistic ranking of groups of demonstrations, without requiring pairwise comparisons over all demonstrations, and 3) adapts to cases where there are varying levels of constraint violation. Our empirical results demonstrate that our proposed Bayesian approach infers constraints of varying severity, more accurately than state-of-the-art constraint inference methods. Code and videos: https://sites.google.com/berkeley.edu/pbicrl.

Reinforcement Learning (RL) has progressed from simple control tasks to complex real-world challenges with large state spaces. While RL excels in these tasks, training time remains a limitation. Reward shaping is a popular solution, but existing methods often rely on value functions, which face scalability issues. This paper presents a novel safety-oriented reward-shaping framework inspired by barrier functions, offering simplicity and ease of implementation across various environments and tasks. To evaluate the effectiveness of the proposed reward formulations, we conduct simulation experiments on CartPole, Ant, and Humanoid environments, along with real-world deployment on the Unitree Go1 quadruped robot. Our results demonstrate that our method leads to 1.4-2.8 times faster convergence and as low as 50-60% actuation effort compared to the vanilla reward. In a sim-to-real experiment with the Go1 robot, we demonstrated better control and dynamics of the bot with our reward framework. We have open-sourced our code at https://github.com/Safe-RL-IISc/barrier_shaping.

Active learning strives to reduce the need for costly data annotation, by repeatedly querying an annotator to label the most informative samples from a pool of unlabeled data, and then training a model from these samples. We identify two problems with existing active learning methods for LiDAR semantic segmentation. First, they overlook the severe class imbalance inherent in LiDAR semantic segmentation datasets. Second, to bootstrap the active learning loop when there is no labeled data available, they train their initial model from randomly selected data samples, leading to low performance. This situation is referred to as the cold start problem. To address these problems we propose BaSAL, a size-balanced warm start active learning model, based on the observation that each object class has a characteristic size. By sampling object clusters according to their size, we can thus create a size-balanced dataset that is also more class-balanced. Furthermore, in contrast to existing information measures like entropy or CoreSet, size-based sampling does not require a pretrained model, thus addressing the cold start problem effectively. Results show that we are able to improve the performance of the initial model by a large margin. Combining warm start and size-balanced sampling with established information measures, our approach achieves comparable performance to training on the entire SemanticKITTI dataset, despite using only 5% of the annotations, outperforming existing active learning methods. We also match the existing state-of-the-art in active learning on nuScenes. Our code is available at: https://github.com/Tony-WJR/BaSAL.

Vision-centric bird-eye-view (BEV) perception has shown promising potential in autonomous driving. Recent works mainly focus on improving efficiency or accuracy but neglect the challenges when facing environment changing, resulting in severe degradation of transfer performance. For BEV perception, we figure out the significant domain gaps existing in typical real-world cross-domain scenarios and comprehensively solve the Domain Adaption (DA) problem for multi-view 3D object detection. Since BEV perception approaches are complicated and contain several components, the domain shift accumulation on multiple geometric spaces (i.e., 2D, 3D Voxel, BEV) makes BEV DA even challenging. In this paper, we propose a Multi-space Alignment Teacher-Student (MATS) framework to ease the domain shift accumulation, which consists of a Depth-Aware Teacher (DAT) and a Geometric-space Aligned Student (GAS) model. DAT tactfully combines target lidar and reliable depth prediction to construct depth-aware information, extracting target domain-specific knowledge in Voxel and BEV feature spaces. It then transfers the sufficient domain knowledge of multiple spaces to the student model. In order to jointly alleviate the domain shift, GAS projects multi-geometric space features to a shared geometric embedding space and decreases data distribution distance between two domains. To verify the effectiveness of our method, we conduct BEV 3D object detection experiments on three cross-domain scenarios and achieve state-of-the-art performance. Code: https://github.com/liujiaming1996/BEVUDA.

Panoptic segmentation is a challenging perception task, which can help robots to comprehensively perceive the surrounding environment. In the task, we notice that semantic, instance, and panoptic have rich relations, however, which are rarely explored. In this work, we propose a novel panoptic, instance, and semantic bridged network to delve into the reciprocal relation. To make semantic and instance benefit from each other, we design a novel Gated Encoding (GE) module, incorporating complementary cues between semantic and instance heads through the gated mechanism. In addition, a novel edge-aware consistency constraint among edges of each task is presented, which exhaustedly exploits geometric constraints, to boost the segmentation quality of challenging edges. Experimental results on the Cityscapes and MS-COCO datasets demonstrate that our approach achieves state-of-the-art performance in an efficient CNN-based paradigm, attaining a balance between accuracy and efficiency.

The use of autonomous underwater vehicles (AUVs) to accomplish traditionally challenging and dangerous tasks has proliferated thanks to advances in sensing, navigation, manipulation, and on-board computing technologies. Utilizing AUVs in underwater human-robot interaction (UHRI) has witnessed comparatively smaller levels of growth due to limitations in bi-directional communication and significant technical hurdles to bridge the gap between analogies with terrestrial interaction strategies and those that are possible in the underwater domain. A necessary component to support UHRI is establishing a system for safe robotic-diver approach to establish face-to-face communication that considers nonstandard human body pose. In this work, we introduce a stereo vision system for enhancing UHRI that utilizes threedimensional reconstruction from stereo image pairs and machine learning for localizing human joint estimates. We then establish a convention for a coordinate system that encodes the direction the human is facing with respect to the camera coordinate frame. This allows automatic setpoint computation that preserves human body scale and can be used as input to an image-based visual servo control scheme. We show that our setpoint computations tend to agree both quantitatively and qualitatively with experimental setpoint baselines. The methodology introduced shows promise for enhancing UHRI by improving robotic perception of human orientation underwater.

A drawing robot avatar is a robotic system that allows for telepresence-based drawing, enabling users to remotely control a robotic arm and create drawings in real-time from a remote location. The proposed control framework aims to improve bimanual robot telepresence quality by reducing the user workload and required prior knowledge through the automation of secondary or auxiliary tasks. The introduced novel method calculates the near-optimal Cartesian end-effector pose in terms of visual feedback quality for the attached eye-to-hand camera with motion constraints in consideration. The effectiveness is demonstrated by conducting user studies of drawing reference shapes using the implemented robot avatar compared to stationary and teleoperated camera pose conditions. Our results demonstrate that the proposed control framework offers improved visual feedback quality and drawing performance.

Aerial unperching of multirotors has received little attention as opposed to perching that has been investigated to elongate operation time. This study presents a new aerial robot capable of both perching and unperching autonomously on/from a ferromagnetic surface during flight, and a switching controller to avoid rotor saturation and mitigate overshoot during transition between free-flight and perching. To enable stable perching and unperching maneuvers on/from a vertical surface, a lightweight (≈ 1 kg), fully actuated tiltrotor that can hover at 90◦ pitch angle is first developed. We design a perching/unperching module composed of a single servomotor and a magnet, which is then mounted on the tiltrotor. A switching controller including exclusive control modes for transitions between free-flight and perching is proposed. Lastly, we propose a simple yet effective strategy to ensure robust perching in the presence of measurement and control errors and avoid collisions with the perching site immediately after unperching. We validate the proposed framework in experiments where the tiltrotor successfully performs perching and unperching on/from a vertical surface during flight. We further show effectiveness of the proposed transition mode in the switching controller by ablation studies where large overshoot and even collision with a perching site occur. To the best of the authors’ knowledge, this work presents the first autonomous aerial unperching framework using a fully actuated tiltrotor.

Environmental monitoring via UAVs offers unprecedented aerial observation capabilities. However, the limited flight durations of typical multirotors and the demands on human attention in outdoor missions call for more autonomous solutions. Addressing the specific challenges of precise UAV landings – especially amidst wind disturbances, obstacles, and unreliable global localization – we introduce a mobile hub concept. This hub facilitates continuous mission cycling for unmodified off-the-shelf UAVs. Our approach centers on a small landing platform affixed to a robotic arm, adeptly correcting UAV pose errors in windy conditions. Compact enough for installation in an economy car, the system emphasizes two novel strategies. Firstly, external visual tracking of the UAV informs the landing controls for both the drone and the robotic arm. The arm compensates for UAV positioning errors and aligns the platform’s attitude with the UAV for stable landings, even on small platforms under windy conditions. Secondly, the robotic arm can transport the UAV inside the hub, perform maintenance tasks like battery replacements, and then facilitate direct relaunches. Importantly, our design places all operational responsibility on the hub, ensuring the UAV remains unaltered. This ensures broad compatibility with standard UAVs, only necessitating an API for attitude setpoints. Experimental results underscore the efficiency of our model, achieving safe landings with minimal errors (≤ 7 cm) in winds up to 5 Beaufort (8.1 m/s). In essence, our mobile hub concept significantly boosts UAV mission availability, allowing for autonomous operations even under challenging conditions.

In this work, we propose a novel strategy to ensure infants, who inadvertently displace their quilts during sleep, are promptly and accurately re-covered. Our approach is formulated into two subsequent steps: interference resolution and quilt spreading. By leveraging the DWPose human skeletal detection and the Segment Anything instance segmentation models, the proposed method can accurately recognize the states of the infant and the quilt over her, which involves addressing the interferences resulted from an infant’s limbs laid on part of the quilt. Building upon prior research, the EM*D deep learning model is employed to forecast quilt state transitions before and after quilt spreading actions. To improve the sensitivity of the network in distinguishing state variation of the handled quilt, we introduce an enhanced loss function that translates the voxelized quilt state into a more representative one. Both simulation and real-world experiments validate the efficacy of our method, in spreading and recover a quilt over an infant.

Gecko-inspired adhesives have the advantage of being able to grasp and release flat surfaces in a vacuum using their microwedge structures. This makes them an especially attractive solution for perching on and grasping flat objects in space for free-flying robots. To grasp and anchor onto these flat surfaces, the gripper must ensure contact between the gecko adhesives and the surface before applying the appropriate forces to activate their adhesion. However, in the case of a free-flying robot in microgravity, physical contact with the surface induces reaction forces, causing the robot to quickly bounce away from the surface. To solve this issue, we propose a simple passive mechanism and a control method of a robotic arm on a free-flying robot with a gecko adhesive gripper. The gripper utilizes a single-motor controlled tendon-driven mechanism mounted at the end of a robotic arm equipped with controllable stiffness joints and a linear spring-damper system. A free-flying robot on an air-bearing platform can successfully perch on a flat surface with a velocity of up to 72.5mm/s and with an approach angle misalignment of up to 33.0 degrees.

We present a fully autonomous self-recharging drone system capable of long-duration sustained operations near powerlines. The drone is equipped with a robust onboard perception and navigation system that enables it to locate powerlines and approach them for landing. A passively actuated gripping mechanism grasps the powerline cable during landing after which a control circuit regulates the magnetic field inside a split-core current transformer to provide sufficient holding force as well as battery recharging.The system is evaluated in an active outdoor three-phase powerline environment. We demonstrate multiple contiguous hours of fully autonomous uninterrupted drone operations composed of several cycles of flying, landing, recharging, and takeoff, validating the capability of extended, essentially unlimited, operational endurance.

We propose a new method for autonomous navigation in uneven terrains by utilizing a sparse Gaussian Process (SGP) based local perception model. The SGP local perception model is trained on local ranging observation (pointcloud) to learn the terrain elevation profile and extract the feasible navigation subgoals around the robot. Subsequently, a cost function, which prioritizes the safety of the robot in terms of keeping the robot’s roll and pitch angles bounded within a specified range, is used to select a safety-aware subgoal that leads the robot to its final destination. The algorithm is designed to run in real-time and is intensively evaluated in simulation and real-world experiments. The results compellingly demonstrate that our proposed algorithm consistently navigates uneven terrains with high efficiency and surpasses the performance of other planners. The implementation of our method, including the supplementary video showing the experimental and real-world results, is available at https://rb.gy/3ov2r8.

Implicit neural representations have demonstrated significant promise for 3D scene reconstruction. Recent works have extended their applications to autonomous implicit reconstruction through the Next Best View (NBV) based method. However, the NBV method cannot guarantee complete scene coverage and often necessitates extensive viewpoint sampling, particularly in complex scenes. In the paper, we propose to 1) incorporate frontier-based exploration tasks for global coverage with implicit surface uncertainty-based reconstruction tasks to achieve high-quality reconstruction. and 2) introduce a method to achieve implicit surface uncertainty using color uncertainty, which reduces the time needed for view selection. Further with these two tasks, we propose an adaptive strategy for switching modes in view path planning, to reduce time and maintain superior reconstruction quality. Our method exhibits the highest reconstruction quality among all planning methods and superior planning efficiency in methods involving reconstruction tasks. We deploy our method on a UAV and the results show that our method can plan multi-task views and reconstruct a scene with high quality.

Robotic systems for manipulation in millimeter scale often use a camera with high magnification for visual feedback of the target region. However, the limited field-of-view (FoV) of the microscopic camera necessitates camera motion to capture a broader workspace environment. In this work, we propose an autonomous robotic control method to constrain a robot-held camera within a designated FoV. Furthermore, we model the camera extrinsics as part of the kinematic model and use camera measurements coupled with a U-Net based tool tracking to adapt the complete robotic model during task execution. As a proof-of-concept demonstration, the proposed framework was evaluated in a bi-manual setup, where the microscopic camera was controlled to view a tool moving in a pre-defined trajectory. The proposed method allowed the camera to stay 94.1% of the time within the real FoV, compared to 54.4% without the proposed adaptive control.

Autonomous exploration using unmanned aerial vehicles (UAVs) is essential for various tasks such as building inspections, rescue operations, deliveries, and warehousing. However, there are two main limitations to previous approaches: they may not be able to provide a complete map of the environment and assume that the map built during exploration is accurate enough for safe navigation, which is usually not the case. To address these limitations, a novel exploration method is proposed that combines frontier-based exploration with a collector strategy that achieves global exploration and complete map creation. In each iteration, the collector strategy stores and validates frontiers detected during exploration and selects the next best frontier to navigate to. The collector strategy ensures global exploration by balancing the exploitation of a known map with the exploration of unknown areas. In addition, the online path replanning ensures safe navigation through the map created during motion. The performance of the proposed method is verified by exploring 3D simulation environments in comparison with the state-of-the-art methods. Finally, the proposed approach is validated in a real-world experiment.