A hybrid approach to fast and accurate localization for legged robots

Robotica ◽  
2008 ◽  
Vol 26 (6) ◽  
pp. 817-830 ◽  
Author(s):  
Renato Samperio ◽  
Huosheng Hu ◽  
Francisco Martín ◽  
Vicente Matellán
Keyword(s):  
Tracking System ◽  
Hybrid Approach ◽  
Ground Truth ◽  
Legged Robots ◽  
Walking Robots ◽  
Localization Method ◽  
Accurate Localization ◽  
Combined Solution ◽  
Different Characteristics ◽  
Sony Aibo

SUMMARYThis paper describes a hybrid approach to a fast and accurate localization method for legged robots based on Fuzzy-Markov (FM) and Extended Kalman Filters (EKF). Both FM and EKF techniques have been used in robot localization and exhibit different characteristics in terms of processing time, convergence, and accuracy. We propose a Fuzzy-Markov–Kalman (FM–EKF) localization method as a combined solution for a poor predictable platform such as Sony Aibo walking robots. The experimental results show the performance of EKF, FM, and FM-EKF in a localization task with simple movements, combined behaviors, and kidnapped situations. An overhead tracking system was adopted to provide a ground truth to verify the performance of the proposed method.

A Paradigm Shift in Monitoring Pipeline Leaks - Combining Sensorless NPW with RTTM

2021 ◽  
Author(s):  
Eran Inbar ◽  
Eitan Rowen ◽  
Avi Motil ◽  
Eitan Elkin ◽  
Michael Tankersley ◽  
...  
Keyword(s):  
Hybrid Approach ◽  
Leak Detection ◽  
Transient Model ◽  
Fiber Sensing ◽  
Accurate Localization ◽  
Physical Measurements ◽  
Novel Approach ◽  
Hybrid Solution ◽  
Combined Solution ◽  
New Applications

Abstract Leak detection solutions in pipelines use several known methods and technologies. However, each method and its underlying technology has their benefits and drawbacks. This article will present and evaluate a hybrid solution that combines two methods based on different physical measurements and quantities to ensure a superior detection probability, short detection time, accurate localization of faults, and minimal false alarm rates. In addition, this solution also features preventive capabilities by pointing out problematic areas in a pipeline that may need more attention. The article presents a novel approach for pipeline monitoring using a combined solution with the strengths of real-time transient model (RTTM) technology and the power of next-generation fiber sensing geared towards leak detection. On top of acoustic sensing for leaks, it features continuous pipeline integrity monitoring where, using subtle characteristics of propagating negative pressure waves (NPW), pipeline sections signatures are tracked, aiming to detect changes that might expose pipeline integrity issues that can enable the operator to take preventive measures and plan maintenance events. Such a hybrid solution, from AVEVA™ (RTTM) and Prisma Photonics (fiber sensing), will obtain higher levels of performance and reliability. In addition, such a hybrid approach responds to the increasing regulatory demand to have two continuously working solutions based on different physical measures to ensure leak detection and prevention of substance spillage. This article intends to introduce such a hybrid solution with new applications in predictive maintenance for pipeline operators and shed more light on the benefits of such a solution facing further regulatory demands.

scholarly journals Experimental Evaluation of Computer Vision and Machine Learning-Based UAV Detection and Ranging

Drones ◽  
2021 ◽  
Vol 5 (2) ◽  
pp. 37
Author(s):  
Bingsheng Wei ◽  
Martin Barczyk
Keyword(s):  
Machine Learning ◽  
Mean Squared Error ◽  
Tracking System ◽  
Ground Truth ◽  
White Background ◽  
Cascade Classifier ◽  
Detection Algorithms ◽  
Squared Error ◽  
Test Conditions ◽  
Video Feed

We consider the problem of vision-based detection and ranging of a target UAV using the video feed from a monocular camera onboard a pursuer UAV. Our previously published work in this area employed a cascade classifier algorithm to locate the target UAV, which was found to perform poorly in complex background scenes. We thus study the replacement of the cascade classifier algorithm with newer machine learning-based object detection algorithms. Five candidate algorithms are implemented and quantitatively tested in terms of their efficiency (measured as frames per second processing rate), accuracy (measured as the root mean squared error between ground truth and detected location), and consistency (measured as mean average precision) in a variety of flight patterns, backgrounds, and test conditions. Assigning relative weights of 20%, 40% and 40% to these three criteria, we find that when flying over a white background, the top three performers are YOLO v2 (76.73 out of 100), Faster RCNN v2 (63.65 out of 100), and Tiny YOLO (59.50 out of 100), while over a realistic background, the top three performers are Faster RCNN v2 (54.35 out of 100, SSD MobileNet v1 (51.68 out of 100) and SSD Inception v2 (50.72 out of 100), leading us to recommend Faster RCNN v2 as the recommended solution. We then provide a roadmap for further work in integrating the object detector into our vision-based UAV tracking system.

An experimental study on feature-based SLAM for multi-legged robots with RGB-D sensors

2017 ◽  
Vol 44 (4) ◽  
pp. 428-441 ◽  
Author(s):  
Michał R. Nowicki ◽  
Dominik Belter ◽  
Aleksander Kostusiak ◽  
Petr Cížek ◽  
Jan Faigl ◽  
...  
Keyword(s):  
Performance Metrics ◽  
State Of The Art ◽  
Building Blocks ◽  
Bundle Adjustment ◽  
The State ◽  
Legged Robots ◽  
Evaluation Methodology ◽  
Walking Robots ◽  
Content Type ◽  
Feature Based

Purpose This paper aims to evaluate four different simultaneous localization and mapping (SLAM) systems in the context of localization of multi-legged walking robots equipped with compact RGB-D sensors. This paper identifies problems related to in-motion data acquisition in a legged robot and evaluates the particular building blocks and concepts applied in contemporary SLAM systems against these problems. The SLAM systems are evaluated on two independent experimental set-ups, applying a well-established methodology and performance metrics. Design/methodology/approach Four feature-based SLAM architectures are evaluated with respect to their suitability for localization of multi-legged walking robots. The evaluation methodology is based on the computation of the absolute trajectory error (ATE) and relative pose error (RPE), which are performance metrics well-established in the robotics community. Four sequences of RGB-D frames acquired in two independent experiments using two different six-legged walking robots are used in the evaluation process. Findings The experiments revealed that the predominant problem characteristics of the legged robots as platforms for SLAM are the abrupt and unpredictable sensor motions, as well as oscillations and vibrations, which corrupt the images captured in-motion. The tested adaptive gait allowed the evaluated SLAM systems to reconstruct proper trajectories. The bundle adjustment-based SLAM systems produced best results, thanks to the use of a map, which enables to establish a large number of constraints for the estimated trajectory. Research limitations/implications The evaluation was performed using indoor mockups of terrain. Experiments in more natural and challenging environments are envisioned as part of future research. Practical implications The lack of accurate self-localization methods is considered as one of the most important limitations of walking robots. Thus, the evaluation of the state-of-the-art SLAM methods on legged platforms may be useful for all researchers working on walking robots’ autonomy and their use in various applications, such as search, security, agriculture and mining. Originality/value The main contribution lies in the integration of the state-of-the-art SLAM methods on walking robots and their thorough experimental evaluation using a well-established methodology. Moreover, a SLAM system designed especially for RGB-D sensors and real-world applications is presented in details.

Feasibility Study of Designing a Three Legged Walking Robot: Tribot

1991 ◽  
Author(s):  
Yueh-Jaw Lin ◽  
Aaron Tegland
Keyword(s):  
Feasibility Study ◽  
Legged Robots ◽  
Walking Robots ◽  
Design Parameters ◽  
Optimized Design ◽  
Motion Simulation ◽  
Walking Robot ◽  
Design Considerations ◽  
Simulation Based ◽  
Weight Distributions

Abstract In recent years, walking robot research has become an important robotic research topic because walking robots possess mobility, as oppose to stationary robots. However, current walking robot research has only concentrated on even numbered legged robots. Walking robots with odd numbered legs are still lack of attention. This paper presents the study on an odd numbered legged (three-legged) walking robot — Tribot. The feasibility of three-legged walking is first investigated using computer simulation based on a scaled down tribot model. The computer display of motion simulation shows that a walking robot with three legs is feasible with a periodic gait. During the course of the feasibility study, the general design of the three-legged robot is also analyzed for various weights, weight distributions, and link lengths. In addition, the optimized design parameters and limitations are found for certain knee arrangements. These design considerations and feasibility study using computer display can serve as a general guideline for designing odd numbered legged robots.

Indian River Watershed Image Analysis Using Fuzzy-CA Hybrid Approach

2017 ◽  
pp. 232-246
Author(s):  
Kalyan Mahata ◽  
Subhasish Das ◽  
Rajib Das ◽  
Anasua Sarkar
Keyword(s):  
Cellular Automata ◽  
Land Cover ◽  
Discrete Dynamical System ◽  
Hybrid Approach ◽  
Ground Truth ◽  
Second Phase ◽  
River Watershed ◽  
Set Membership ◽  
Mixed Pixels ◽  
Indian River

Image segmentation among overlapping land cover areas in satellite images is a very crucial task. Detection of belongingness is the important problem for classifying mixed pixels. This paper proposes an approach for pixel classification using a hybrid approach of Fuzzy C-Means and Cellular automata methods. This new unsupervised method is able to detect clusters using 2-Dimensional Cellular Automata model based on fuzzy segmentations. This approach detects the overlapping regions in remote sensing images by uncertainties using fuzzy set membership values. As a discrete, dynamical system, cellular automaton explores uniformly interconnected cells with states. In the second phase of our method, we utilize a 2-dimensional cellular automata to prioritize allocations of mixed pixels among overlapping land cover areas. We experiment our method on Indian Ajoy river watershed area. The clustered regions are compared with well-known FCM and K-Means methods and also with the ground truth knowledge. The results show the superiority of our new method.

Locomotion Interfaces for Legged Robots

2020 ◽  
pp. 246-287
Author(s):  
Hisham A. Abdel-Aal
Keyword(s):  
Comparative Study ◽  
Pressure Distribution ◽  
Design Process ◽  
Structural Integrity ◽  
Close Correlation ◽  
Legged Robots ◽  
Walking Robots ◽  
Textural Parameters

This chapter presents a comparative study of the topographical structure of three common biological robotic inspirations: human, canine, and feline feet. It is shown that the metrological roughness of each of the examined feet is customized for the specific locomotion demands of the species. The textural parameters manifest close correlation to the pressure distribution experienced in movement and gait. This correlation enhances the durability and structural integrity of the bio-analogue. It is also shown that the metrological function of the human (plantigrade) feet pads combine that of the back and the front feet pads of the digitigrade mammals examined. It is argued that integrating the targeted engineering of roughness within the design process of robotic feet can enhance the function of walking robots. Further, it offers elegant solutions to some of the current problems encountered in design of humanoids and other bio-inspired walking robots.

scholarly journals A Hybrid Deep Learning Framework for Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data

2020 ◽  
Vol 10 (15) ◽  
pp. 5191
Author(s):  
Yıldız Karadayı ◽  
Mehmet N. Aydin ◽  
A. Selçuk Öğrenci
Keyword(s):  
Deep Learning ◽  
Anomaly Detection ◽  
Hurricane Katrina ◽  
Time Series Data ◽  
Hybrid Approach ◽  
Ground Truth ◽  
Outbreak Detection ◽  
Series Data ◽  
Detection Techniques ◽  
Spatio Temporal

Multivariate time-series data with a contextual spatial attribute have extensive use for finding anomalous patterns in a wide variety of application domains such as earth science, hurricane tracking, fraud, and disease outbreak detection. In most settings, spatial context is often expressed in terms of ZIP code or region coordinates such as latitude and longitude. However, traditional anomaly detection techniques cannot handle more than one contextual attribute in a unified way. In this paper, a new hybrid approach based on deep learning is proposed to solve the anomaly detection problem in multivariate spatio-temporal dataset. It works under the assumption that no prior knowledge about the dataset and anomalies are available. The architecture of the proposed hybrid framework is based on an autoencoder scheme, and it is more efficient in extracting features from the spatio-temporal multivariate datasets compared to the traditional spatio-temporal anomaly detection techniques. We conducted extensive experiments using buoy data of 2005 from National Data Buoy Center and Hurricane Katrina as ground truth. Experiments demonstrate that the proposed model achieves more than 10% improvement in accuracy over the methods used in the comparison where our model jointly processes the spatial and temporal dimensions of the contextual data to extract features for anomaly detection.

Accurate localization method for subaperture stitching interferometry in aspherical optics metrology

2020 ◽  
Vol 91 (7) ◽  
pp. 075114
Author(s):  
Zhuo Zhao ◽  
Bing Li ◽  
Xiaoqin Kang ◽  
Jiasheng Lu ◽  
Xiang Wei ◽  
...  
Keyword(s):  
Localization Method ◽  
Accurate Localization ◽  
Subaperture Stitching ◽  
Aspherical Optics

scholarly journals Framework for Developing Bio-Inspired Morphologies for Walking Robots

2020 ◽  
Vol 10 (19) ◽  
pp. 6986
Author(s):  
Peter Billeschou ◽  
Nienke N. Bijma ◽  
Leon B. Larsen ◽  
Stanislav N. Gorb ◽  
Jørgen C. Larsen ◽  
...  
Keyword(s):  
Walking Speed ◽  
Dung Beetle ◽  
Object Manipulation ◽  
Step Length ◽  
Legged Robots ◽  
Walking Robots ◽  
Beetle Species ◽  
Step Frequency ◽  
Development Framework ◽  
Interdisciplinary Collaborations

Morphology is a defining trait of any walking entity, animal or robot, and is crucial in obtaining movement versatility, dexterity and durability. Collaborations between biologist and engineers create opportunities for implementing bio-inspired morphologies in walking robots. However, there is little guidance for such interdisciplinary collaborations and what tools to use. We propose a development framework for transferring animal morphologies to robots and substantiate it with a replication of the ability of the dung beetle species Scarabaeus galenus to use the same morphology for both locomotion and object manipulation. As such, we demonstrate the advantages of a bio-inspired dung beetle-like robot, ALPHA, and how its morphology outperforms a conventional hexapod by increasing the (1) step length by 50.0%, (2) forward and upward reach by 95.5%, and by lowering the (3) overall motor acceleration by 7.9%, and (4) step frequency by 21.1% at the same walking speed. Thereby, the bio-inspired robot has longer and fewer steps that lower fatigue-inducing impulses, a greater variety of step patterns, and can potentially better utilise its workspace to overcome obstacles. Hence, we demonstrate how the framework can be used to develop legged robots with bio-inspired morphologies that embody greater movement versatility, dexterity and durability.

scholarly journals AGV Localization System Based on Ultra-Wideband and Vision Guidance

Electronics ◽  
2020 ◽  
Vol 9 (3) ◽  
pp. 448 ◽  
Author(s):  
Xiaohao Hu ◽  
Zai Luo ◽  
Wensong Jiang
Keyword(s):  
Monocular Vision ◽  
Ultra Wideband ◽  
Localization Algorithm ◽  
Localization Accuracy ◽  
Localization Method ◽  
Localization System ◽  
Indoor Location ◽  
Accurate Localization ◽  
Vision Localization ◽  
Automated Storage

Aiming at the problems of low localization accuracy and complicated localization methods of the automatic guided vehicle (AGV) in the current automatic storage and transportation process, a combined localization method based on the ultra-wideband (UWB) and the visual guidance is proposed. Both the UWB localization method and the monocular vision localization method are applied to the indoor location of the AGV. According to the corner points of an ArUco code fixed on the AGV body, the monocular vision localization method can solve the pose information of the AGV by the PnP algorithm in real-time. As an auxiliary localization method, the UWB localization method is called to locate the AGV coordinates. The distance from the tag on the AGV body to the surrounding anchors is measured by the time of flight (TOF) ranging algorithm, and the actual coordinates of the AGV are calculated by the trilateral centroid localization algorithm. Then, the localization data of the UWB is corrected by the mean compensation method to obtain a consistent and accurate localization trajectory. The experiment result shows that this localization system has an error of 15mm, which meets the needs of AGV location in the process of automated storage and transportation.

Sign in / Sign up

Export Citation Format

Share Document