With the widespread use of distributed systems, multi-subspace whole-flow industrial monitoring methods are evolving. However, due to the lack of distinctive features, incipient faults in plant-wide processes are more difficult to detect. To improve the detection rate of incipient faults in plant-wide processes while maintaining the generality of the algorithm, a novel double-layer subspace weighted moving window reconstruction independent component analysis (DS-WRICA) method is proposed. In DS-WRICA, process variables are first divided into different subspaces based on process knowledge and data-driven partitioning methods. Secondly, a weighted moving window is used to increase the offset of incipient faults, and monitoring statistics are constructed by combining optimized reconstructed independent component analysis (RICA) and local outlier factor (LOF) in each subspace. Then, the monitoring statistics in each subspace are fused with information by Bayesian inference fusion method to obtain distributed monitoring results. Finally, the effectiveness and superiority of the DS-WRICA method are verified by industrial examples.
Hexapod robots used in resource exploration and post-disaster rescue operations often face the risk of multiple unexpected leg failures, which can lead to an immediate threat of falling. Traditional controllers generally lack the fault-tolerance required to adjust locomotion gaits to continue to move in a target direction. Maintaining the original motion patterns in such situations can result in severe damage to the robot. This work proposes a hierarchical fault-tolerant control scheme that employs a meta-learning training architecture capable of addressing the non-stationary multi-task leg failure adaptation problem. The meta Learning architecture simultaneously trains multiple locomotion tasks with all possible configurations of one, two or three leg failures occurring during the normal movement of hexapod robots. Experimental results demonstrate the exceptional fault-tolerant capabilities of the trained controller in adapting to unexpected leg failures. Furthermore, the controller maintains the robot’s planar mobility, providing superior velocity tracking and directional following abilities compared to baselines.
The ultrasound robotic system (URS) is of tremendous importance for assisting sonographers to diagnose various diseases. Generally, the quality of ultrasound image which is evaluated through confidence map has a high dependency on the experience of sonographers. In order to obtain high-quality ultrasound image under URS, we take confidence map into consideration of control frame and propose confidence-driven adaptive optimal impedance learning scheme. By introducing an appropriate performance function which blends both confidence value and contacting force, an iterative learning strategy is conducted to achieve favorable scanning performance. Finally, we show the experiment results on the human volunteer.
Over the years, significant developments and applications of light detection and ranging (LiDAR)-based simultaneous localization and mapping (SLAM) have occurred in robotic mapping, in part due to the high precision and decreasing costs of LiDAR. This study overviews the most advanced spinning-actuated LiDAR-based SLAM systems and discusses the challenges of meeting modern robotic mapping needs. In general, we intend to provide a substantial resource for researchers and engineers involved in robotic mapping.
Last-mile delivery robot has been attracted increasing attention from industry and comes into our daily life recently. However, how to safely and effectively navigate among crowded pedestrians is still an open problem. It requires the robot capable of analysing where it can traverse, understanding the intentions of surrounding pedestrians, planning the trajectory with social awareness, etc. In this paper, we have successfully completed a systematic implementation for navigation of delivery robot in pedestrian crowded environments. First, we introduced the Nanyang Sidewalk dataset, designed explicitly for class segmentation tasks on sidewalks. Second, a multi-modal 3D detection and motion prediction integrated with the social force model has been introduced to perceive the intention of pedestrians. Then, a socially aware motion planner for the delivery robot is demonstrated by following pedestrian etiquette. Extensive experiments have been conducted to verify and evaluate the performance of the proposed algorithm.
