Journal of Field Robotics最新文献

Planetary rovers may become stuck due to the soft terrain on Mars and other planetary surface. The escape entrapment control strategy is of great significance for planetary rover traversing loosely consolidated granular terrain. After analyzing the performance of the published quadrupedal rotary sequence gait, a “sweeping-spinning” gait was proposed to improve escape entrapment capability. And the forward distance of planetary rovers with “sweeping-spinning” gait was modeled as a function of six control parameters. An online optimization escape entrapment strategy for planetary rover was proposed based on the Bayesian Optimization algorithm. Single-factor experiments were conducted to investigate the effect of each control parameter on forward distance, and determine the parameter ranges. The average forward distance with randomly selected control parameters is 89.64 cm, while that is 136.93 cm with Bayesian optimized control parameters, which verifies the effectiveness of the escape entrapment strategy. Moreover, compared with the trajectory of a planetary rover prototype with the published quadrupedal rotary sequence gait, the trajectory of a planetary rover prototype with “sweeping-spinning” gait is more accurate. Furthermore, the online estimated equivalent terrain mechanical parameters can be used to determine the running state of the planetary rover prototype, which was verified using experiments.

Focused on the problems of cumbersome operation, low efficiency, and high cost in the traditional manual rebar binding process, we propose a mobile robot vision detection and path-planning method for rebar binding to realize automated rebar binding by combining deep learning and path-planning technology. A MobileNetV3-SSD rebar binding crosspoints recognition model is built based on TensorFlow deep learning framework, and a crosspoints localization method combining control factor α and feature projection curve is introduced to achieve the localization of unbound crosspoints. In addition, A back-and-forth path-planning algorithm with priority constraints combined with dead zone escape algorithm based on improved A* is proposed to achieve complete coverage path planning of the working area and path transfer of the dead zone. In the field test of the robot prototype, the classification accuracy and localization accuracy reached 94.40% and 90.49%, and the robot was able to reach complete coverage path planning successfully. The experimental results show that the visual detection method can achieve fast, noncontact and intelligent recognition of rebar binding crosspoints, which has good robustness and application value. At the same time, the proposed path-planning method has higher efficiency in the execution of robot complete coverage path planning, and meets the basic requirements of path planning for rebar binding process.

With the widespread adoption of mobile robots, effective path planning has become increasingly critical. Although traditional search methods have been extensively utilized, meta-heuristic algorithms have gained popularity owing to their efficiency and problem-specific heuristics. However, challenges remain in terms of premature convergence and lack of solution diversity. To address these issues, this paper proposes a novel artificial potential field enhanced improved multiobjective snake optimization algorithm (APF-IMOSO). This paper presents four key enhancements to the snake optimizer to significantly improve its performance. Additionally, it introduces four fitness functions focused on optimizing path length, safety (evaluated via artificial potential field method), energy consumption, and time efficiency. The results of simulation and experiment in four scenarios including static and dynamic highlight APF-IMOSO's advantages, delivering improvements of 8.02%, 7.61%, 50.71%, and 12.74% in path length, safety, energy efficiency, and time-savings, respectively, over the original snake optimization algorithm. Compared with other advanced meta-heuristics, APF-IMOSO also excels in these indexes. Real robot experiments show an average path length error of 1.19% across four scenarios. The results reveal that APF-IMOSO can generate multiple viable collision-free paths in complex environments under various constraints, showcasing its potential for use in dynamic path planning within the realm of robot navigation.

The cover image is based on the Research Article The Haidou-1 hybrid underwater vehicle for the Mariana Trench science exploration to 10,908 m depth by Jian Wang et al., https://doi.org/10.1002/rob.22307

Litter nowadays presents a significant threat to the equilibrium of many ecosystems. An example is the sea, where litter coming from coasts and cities via gutters, streets, and waterways, releases toxic chemicals and microplastics during its decomposition. Litter removal is often carried out manually by humans, which inherently lowers the amount of waste that can be effectively collected from the environment. In this paper, we present a novel quadruped robot prototype that, thanks to its natural mobility, is able to collect cigarette butts (CBs) autonomously, the second most common undisposed waste worldwide, in terrains that are hard to reach for wheeled and tracked robots. The core of our approach is a convolutional neural network for litter detection, followed by a time-optimal planner for reducing the time needed to collect all the target objects. Precise litter removal is then performed by a visual-servoing procedure which drives the nozzle of a vacuum cleaner that is attached to one of the robot legs on top of the detected CB. As a result of this particular position of the nozzle, we are able to perform the collection task without even stopping the robot's motion, thus greatly increasing the time-efficiency of the entire procedure. Extensive tests were conducted in six different outdoor scenarios to show the performance of our prototype and method. To the best knowledge of the authors, this is the first time that such a design and method was presented and successfully tested on a legged robot.

This paper fuses ideas from reinforcement learning (RL), Learning from Demonstration (LfD), and Ensemble Learning into a single paradigm. Knowledge from a mixture of control algorithms (experts) are used to constrain the action space of the agent, enabling faster RL refining of a control policy, by avoiding unnecessary explorative actions. Domain-specific knowledge of each expert is exploited. However, the resulting policy is robust against errors of individual experts, since it is refined by a RL reward function without copying any particular demonstration. Our method has the potential to supplement existing RLfD methods when multiple algorithmic approaches are available to function as experts, specifically in tasks involving continuous action spaces. We illustrate our method in the context of a visual servoing (VS) task, in which a 7-DoF robot arm is controlled to maintain a desired pose relative to a target object. We explore four methods for bounding the actions of the RL agent during training. These methods include using a hypercube and convex hull with modified loss functions, ignoring actions outside the convex hull, and projecting actions onto the convex hull. We compare the training progress of each method using expert demonstrators, employing one expert demonstrator with the DAgger algorithm, and without using any demonstrators. Our experiments show that using the convex hull with a modified loss function not only accelerates learning but also provides the most optimal solution compared with other approaches. Furthermore, we demonstrate faster VS error convergence while maintaining higher manipulability of the arm, compared with classical image-based VS, position-based VS, and hybrid-decoupled VS.

Learning-based visual odometry (VO) becomes popular as it achieves a remarkable performance without manually crafted image processing and burdensome calibration. Meanwhile, the inertial navigation can provide a localization solution to assist VO when the VO produces poor state estimation under challenging visual conditions. Therefore, the combination of learning-based technique and classical state estimation method can further improve the performance of pose estimation. In this paper, we propose a learning-based visual-inertial odometry (VIO) algorithm, which consists of an end-to-end VO network and an $��\; ��S�\; �E_2�\; ��\left(�\; �3�\; �\right)���$-Extended Kalman Filter (EKF). The VO network mainly combines a convolutional neural network with a recurrent neural network, taking advantage of two consecutive monocular images to produce relative pose estimation with associated uncertainties. The $��\; ��S�\; �E_2�\; ��\left(�\; �3�\; �\right)���$-EKF, which is proved to overcome the inconsistency issues of VIO, propagates inertial measurement unit kinematics-based states, and fuses relative measurements and uncertainties from the VO network in its update step. The extensive experimental results on the KITTI data set and the EuRoC data set demonstrate the superior performance of the proposed method compared to other related methods.

This paper presents the design and development of a funnel-shaped Sparus Docking Station intended for the non-holonomic torpedo-shaped Sparus II Autonomous Underwater Vehicle. The Sparus Docking Station is equipped with sensors and batteries, allowing for a stand-alone long-term deployment of the vehicle. An inverted Ultra Short Base-Line system is used to locate the Docking Station as well as to provide long-term drift-less vehicle navigation. The Sparus Docking Station is able to observe the ocean currents using a Doppler Velocity Log, being motorized to allow its self-alignment with the current. Moreover, a docking algorithm accounting for the current is used to guide the robot during the docking maneuver. The paper reports consecutive successful experimental results of the docking maneuver in sea trials in two different countries.

Plastering is dominated manually, exhibiting low levels of automation and inconsistent finished quality. A comprehensive review of literature indicates that extant plastering robots demonstrate a subpar performance when tasked with rectifying defects in the transition area. The limitations encompass a lack of capacity to independently evaluate the quality of work or perform remedial plastering procedures. To address this issue, this research describes the system design of the Puttybot and a paradigm of plastering to solve the stated problems. The Puttybot consists of a mobile chassis, a lift platform, and a macro/micromanipulator. The force-controlled scraper parameters have been calibrated to dynamically modify their rigidity in response to the applied putty. This strategy utilizes convolutional neural networks to identify plastering defects and executes the plastering operation with force feedback. This paradigm's effectiveness was validated during an autonomous plastering trial wherein a large-scale wall was processed without human involvement.