This paper studies the application of Deep Q-Networks (DQN) for shortest-path planning on mobile robots. Implementing DQN on mobile robots poses challenges due to the limited computational resources of embedded system...
详细信息
ISBN:
(纸本)9798350367331;9798350367348
This paper studies the application of Deep Q-Networks (DQN) for shortest-path planning on mobile robots. Implementing DQN on mobile robots poses challenges due to the limited computational resources of embedded systems. The process consists of two mode: the training mode, where an inference model of DQN with weight and bias parameters is generated, and the operational mode, where the DQN model is loaded onto the robot to perform actions in a simulation maze environment with gazebo. This study investigates optimal computational techniques for matrix operations, which are the primary operations in DQN. Computational methods performed on embedded system platforms utilizing GPUs (Jetson Xavier NX) involve studying matrix computation techniques using arrays, 2D arrays with GPUs, and tensors. Based on the experiment, the results show that for the DQN model used in shortest-path planning, computation based on an array matrix significantly speeds up the process.
Today's robot policies exhibit subpar performance when faced with the challenge of generalizing to novel environments. Human corrective feedback is a crucial form of guidance to enable such generalization. However...
详细信息
ISBN:
(纸本)9798350384581;9798350384574
Today's robot policies exhibit subpar performance when faced with the challenge of generalizing to novel environments. Human corrective feedback is a crucial form of guidance to enable such generalization. However, adapting to and learning from online human corrections is a non-trivial endeavor: not only do robots need to remember human feedback over time to retrieve the right information in new settings and reduce the intervention rate, but also they would need to be able to respond to feedback that can be arbitrary corrections about high-level human preferences to low-level adjustments to skill parameters. In this work, we present Distillation and Retrieval of Online Corrections (DROC), a large language model (LLM)-based system that can respond to arbitrary forms of language feedback, distill generalizable knowledge from corrections, and retrieve relevant past experiences based on textual and visual similarity for improving performance in novel settings. DROC is able to respond to a sequence of online language corrections that address failures in both high-level task plans and low-level skill primitives. We demonstrate that DROC effectively distills the relevant information from the sequence of online corrections in a knowledge base and retrieves that knowledge in settings with new task or object instances. DROC outperforms other techniques that directly generate robot code via LLMs [1] by using only half of the total number of corrections needed in the first round and requires little to no corrections after two iterations. We show further results and videos on our project website: https://***/***/droc.
Although the importance of robots for efficient production in smart factories and assistance in home settings is well recognized, their usage in industry and society still faces challenges. A cause lies in the difficu...
详细信息
ISBN:
(纸本)9798350362923;9798350362916
Although the importance of robots for efficient production in smart factories and assistance in home settings is well recognized, their usage in industry and society still faces challenges. A cause lies in the difficulty that comes with programmingrobots by writing several lines of code to complete basic tasks. Unfortunately, workers in particular as well as citizens in general rarely have the necessary technical background and experience to achieve this complex goal. Recruiting professional robot programmers creates a dependency and increases costs associated with production and assistance. This issue gets more complicated and urgent when robots are deployed to accommodate small batch sizes and heterogeneous preferences from Industry 4.0 and Industry 5.0 applications. In fact, the duration allocated to the changeover in robotized applications might be brief to meet concurrent economic objectives. Furthermore, generic motion primitives provided by robot vendors might not match with irregular and domain-specific Cartesian trajectories in industry, society, and in-between. This work embraces these issues by integrating extended reality and digital twins to propose an intuitive, inclusive, fast, and effective framework to program (i.e., plan and execute) robot motions. Novices can thereby move a manipulator to reach freely specified Cartesian frames and complete grasping maneuvers. This is done without writing any line of code. How our approach empowers citizens is shown by completing manipulations to grasp and relocate workpieces in practice.
Just as humans can become disoriented in featureless deserts or thick fogs, not all environments are conducive to the Localization Accuracy and Stability (LAS) of autonomous robots. This paper introduces an efficient ...
详细信息
ISBN:
(纸本)9798350377712;9798350377705
Just as humans can become disoriented in featureless deserts or thick fogs, not all environments are conducive to the Localization Accuracy and Stability (LAS) of autonomous robots. This paper introduces an efficient framework designed to enhance LiDAR-based LAS through strategic trajectory generation, known as Perception-aware Planning. Unlike vision-based frameworks, the LiDAR-based requires different considerations due to unique sensor attributes. Our approach focuses on two main aspects: firstly, assessing the impact of LiDAR observations on LAS. We introduce a perturbation-induced metric to provide a comprehensive and reliable evaluation of LiDAR observations. Secondly, we aim to improve motion planning efficiency. By creating a Static Observation Loss Map (SOLM) as an intermediary, we logically separate the time-intensive evaluation and motion planning phases, significantly boosting the planning process. In the experimental section, we demonstrate the effectiveness of the proposed metrics across various scenes and the feature of trajectories guided by different metrics. Ultimately, our framework is tested in a real-world scenario, enabling the robot to actively choose topologies and orientations preferable for localization. The source code is accessible at https://***/ZJU-FAST-Lab/LF-3PM.
Assembly planning is a fundamental problem in robotics and automation, which involves designing a sequence of motions to bring the separate constituent parts of a product into their final placement in the product. Ass...
详细信息
The paper presents an approach for interactive programming of the robotic manipulator using mixed reality. The developed system is based on the HoloLens glasses connected through robotic Operation System to Unity engi...
详细信息
ISBN:
(纸本)9781728173955
The paper presents an approach for interactive programming of the robotic manipulator using mixed reality. The developed system is based on the HoloLens glasses connected through robotic Operation System to Unity engine and robotic manipulators. The system gives a possibility to recognize the real robot location by the point cloud analysis, to use virtual markers and menus for the task creation, to generate a trajectory for execution in the simulator or on the real manipulator. It also provides the possibility of scaling virtual and real worlds for more accurate planning. The proposed framework has been tested on pick-and-place and contact operations execution by UR10e and KUKA iiwa robots.
This paper addresses the path-planning challenge for very large-scale robotic systems (VLSR) operating in complex and cluttered environments. VLSR systems consist of numerous cooperative agents or robots working toget...
详细信息
ISBN:
(纸本)9798350382662;9798350382655
This paper addresses the path-planning challenge for very large-scale robotic systems (VLSR) operating in complex and cluttered environments. VLSR systems consist of numerous cooperative agents or robots working together autonomously. Traditionally, many approaches for VLSR systems are developed based on Gaussian mixture models (GMMs), where the GMMs represent agents' evolving spatial distribution, serving as a macroscopic view of the system's state. However, our recent research into VLSR systems has unveiled limitations in using GMMs to represent agent distributions, especially in cluttered environments. To overcome these limitations, we propose a novel model called the skew-normal mixture model (SNMM) for representing agent distributions. Additionally, we present a parameter learning algorithm designed to estimate the SNMM's parameters using sample data. Furthermore, we develop two SNMM-based path-planning algorithms to guide VLSR systems through complex and cluttered environments. Our simulation results demonstrate the effectiveness and superiority of these algorithms compared to GMM-based path-planning methods.
In a robotic sanding operation, three critical aspects require attention: precise identification of the sanding locations, accurate calculation of the necessary sanding amount, and strategic planning for the robot'...
详细信息
ISBN:
(纸本)9798350358513;9798350358520
In a robotic sanding operation, three critical aspects require attention: precise identification of the sanding locations, accurate calculation of the necessary sanding amount, and strategic planning for the robot's sanding trajectory. We introduce a purely mathematical approach to achieve exceptionally precise detection of uneven surfaces under the millimeter level. Our approach entails capturing point cloud images from both sanded and unsanded objects. A best-fit method is employed to enhance the accuracy of Iterative Closest Point (ICP) method matching. The ICP method is then applied iteratively until a high level of accuracy is achieved. By implementing the Angle Criterion method the edge noise in the point cloud, which can arise due to misalignment, are filtered out. The point cloud data representing uneven surfaces is transformed into the robot's base coordinate system to facilitate the preparation of the robotic sanding path planning. The concept was validated on a ceramic wash basin using a 3D camera, a laser line sensor, and a sensitive robot with the ability to perform force-controlled applications like sanding.
Understanding and following natural language instructions while navigating through complex, real-world environments poses a significant challenge for general-purpose robots. These environments often include obstacles ...
详细信息
ISBN:
(纸本)9798891760998
Understanding and following natural language instructions while navigating through complex, real-world environments poses a significant challenge for general-purpose robots. These environments often include obstacles and pedestrians, making it essential for autonomous agents to possess the capability of self-corrected planning to adjust their actions based on feedback from the surroundings. However, the majority of existing vision-andlanguage navigation (VLN) methods primarily operate in less realistic simulator settings and do not incorporate environmental feedback into their decision-making processes. To address this gap, we introduce a novel zero-shot framework called CorNav, utilizing a large language model for decision-making and comprising two key components: 1) incorporating environmental feedback for refining future plans and adjusting its actions, and 2) multiple domain experts for parsing instructions, scene understanding, and refining predicted actions. In addition to the framework, we develop a 3D simulator that renders realistic scenarios using Unreal Engine 5. To evaluate the effectiveness and generalization of navigation agents in a zero-shot multi-task setting, we create a benchmark called NavBench. Our empirical study involves deploying 7 baselines across four tasks, i.e., goal-conditioned navigation given a specific object category, goal-conditioned navigation given simple instructions, finding abstract objects based on high-level instructions, and step-by-step instruction following. Extensive experiments demonstrate that CorNav consistently outperforms all baselines by a significant margin across all tasks.
Accurately assessing the potential value of new sensor observations is a critical aspect of planning for active perception. This task is particularly challenging when reasoning about high-level scene understanding usi...
详细信息
ISBN:
(纸本)9798350384581;9798350384574
Accurately assessing the potential value of new sensor observations is a critical aspect of planning for active perception. This task is particularly challenging when reasoning about high-level scene understanding using measurements from vision-based neural networks. Due to appearance-based reasoning, the measurements are susceptible to several environmental effects such as the presence of occluders, variations in lighting conditions, and redundancy of information due to similarity in appearance between nearby viewpoints. To address this, we propose a new active perception framework incorporating an arbitrary number of perceptual effects in planning and fusion. Our method models the correlation with the environment by a set of general functions termed perceptual factors to construct a perceptual map, which quantifies the aggregated influence of the environment on candidate viewpoints. This information is seamlessly incorporated into the planning and fusion processes by adjusting the uncertainty associated with measurements to weigh their contributions. We evaluate our perceptual maps in a simulated environment that reproduces environmental conditions common in robotics applications. Our results show that, by accounting for environmental effects within our perceptual maps, we improve the state estimation by correctly selecting the viewpoints and considering the measurement noise correctly when affected by environmental factors. We furthermore deploy our approach on a ground robot to showcase its applicability for real-world active perception missions.
暂无评论