Spain

Delayed and intermittent communication with rovers in long-distance missions remains a key challenge for teleoperation, as it affects critically either the liveness or the safety of the rover. This position paper proposes the use of the newly-introduced human-machine interaction paradigm of Machine Coaching, as a middle-ground approach between a fully remotely-controlled and a fully autonomously-operated rover. According to this paradigm, the rover makes decisions and takes actions based on a policy, under the in-situ (even if delayed) supervision of a human coach. Whenever the human coach observes an unexpected decision made by the rover, the former enquires about the reasoning of the latter in making that decision, and then provides feedback on why a possibly different decision (including that of waiting to receive instructions from a remote operator) would have been preferable. This particular protocol, through the bilateral exchange of explanations between the rover and the human coach, allows the rover to provably and efficiently converge towards a policy that is acceptable to the human coach, while also endowing the rover with conditional autonomy in those situations where its current policy offers, in the human coach&#39;s expert opinion, an appropriate balance between liveness and safety. The rover&#39;s initial policy can be designed either in a simulated environment during the rover&#39;s development, or it can be based on early information on the remote environment transmitted once the rover is deployed. 

ECAI 2020

Rover Teleoperation through Machine Coaching

ECAI2020 featured a full programme of technical papers, as well as many AI-related events of interest for researchers, students and all attendants who are interested on contemporary AI, including:




**An extensive programme of workshops, tutorials, and invited speakers**

**The Prestigious Applications of Intelligent Systems conference (PAIS)**

**The Frontiers in AI track sessions**

**The Lab To Market event**

**The Women in AI time**

**The EU Challenges forum**

… and many more!

ECAI 2020 featured a full programme of technical papers, as well as many AI-related events of interest for researchers, students and all attendants who are interested on contemporary AI.

Delayed and intermittent communication with rovers in long-distance missions remains a key challenge for teleoperation, as it affects critically either the liveness or the safety of the rover. This position paper proposes the use of the newly-introduced human-machine interaction paradigm of Machine Coaching, as a middle-ground approach between a fully remotely-controlled and a fully autonomously-operated rover. According to this paradigm, the rover makes decisions and takes actions based on a policy, under the in-situ (even if delayed) supervision of a human coach. Whenever the human coach observes an unexpected decision made by the rover, the former enquires about the reasoning of the latter in making that decision, and then provides feedback on why a possibly different decision (including that of waiting to receive instructions from a remote operator) would have been preferable. This particular protocol, through the bilateral exchange of explanations between the rover and the human coach, allows the rover to provably and efficiently converge towards a policy that is acceptable to the human coach, while also endowing the rover with conditional autonomy in those situations where its current policy offers, in the human coach's expert opinion, an appropriate balance between liveness and safety. The rover's initial policy can be designed either in a simulated environment during the rover's development, or it can be based on early information on the remote environment transmitted once the rover is deployed. 

technical paper

"The dynamics of space objects in Low Earth Orbit are strongly determined by the effects of atmospheric drag. This complex interaction is considered one of the most dominant sources of uncertainty at these altitudes, and is characterised in part by the atmospheric density, a quantity highly correlated to space weather. Current atmosphere models typically account for this via proxy indices such as F10.7, but with variations in solar flux forecasts leading to orbit differences of multiple kilometres over just a few days, prediction of these quantities may represent one of the largest limiting factors in drag modelling. This has fundamental implications both in the short term, in the day-to-day management of operational spacecraft, and in the mid-to-long term, in determining satellite orbital lifetime.
In this work, we apply state-of-the-art time series forecasting techniques and deep learning architectures for the prediction of the F10.7 proxy on the days-ahead timescales relevant to space operations. The novel N-BEATS architecture is employed, which has been shown to achieve winning accuracies in the M4 forecasting competition, and is compared to operationally available forecasts provided by the ESA Space Weather Service Network.

Key words: Deep Learning, Time Series Forecasting, Space Weather, Space Debris
"

A Deep Learning Approach to Space Weather Forecasting for Orbital Prediction

The abundant availability of modern satellite imagery (including optical images and Synthetic Aperture Radar (SAR) data) allows a continuous monitoring of processes on the Earth’s surface. However, training automatic systems to analyze such data requires large amounts of training samples. This is rarely the case for Earth Observation tasks. This talk discusses an example of self-supervised learning that allows to drastically reduce the amount of training data while increasing the achieved performance. The core idea is that learning the proxy task of transcoding SAR images into optical images forces an employed conditional generative adversarial network (GAN) to implicitly distinguish between different land surfaces. Such a network can then be used to build a classifier with significantly fewer free parameters that generalizes well even when trained on a very small amount of labeled data. We train such a GAN on aligned Sentinel-1 and Sentinel-2 image pairs and then show that a pre-trained classifier using these features learned from transcoding outperforms classifiers that are trained from scratch in particular when only a very limited amount of labeled pixels are available for training.

Transcoding-based self-supervised learning for semantic segmentation of PolSAR imagery

"Machine Learning techniques are widely used in the Classification and Segmentation of Satellite
Imagery, however, fewer works are carried out in the field of Spatial Interpolation. The research
paper explains the concept of spatial interpolation and how to carry out spatial interpolation
of a continuous variable using Geostatistical and Machine Learning techniques for a particular
geographical area. Spatial Interpolation is the process of estimating the value of continuous
target variable at unsampled locations based on the values at known sample locations whereas
Supervised Machine Learning is a process of learning a function that maps an input to an
output based on training input features and output labels. Air Temperature is an important
climatic variable that plays a crucial role in global change studies thus accurate estimation of air
temperature is important in several environmental studies. Usually, air temperature is measured at
Automatic Weather Stations (AWS) at different heights and these provide very limited information
about spatial patterns at regional or global scales because they are less in number. Hence, spatial
interpolation of daily air temperature is carried out for Uttarakhand state in India to predict the
values at unsampled locations and the results of geostatistical and machine learning techniques
were compared. Daily temperature from 16 Automatic Weather Stations located in different parts
of Uttarakhand state is considered as the primary dataset for Geostatistical techniques. MODIS
Land Surface Temperature (LST), Normalized Difference Vegetation Index (NDVI), Digital Elevation
Model (DEM), Distance from the nearest Water Body datasets are considered as input features
and the daily temperature is considered as output label for the Machine Learning based approach.
The concept of spatial interpolation was preserved during the training of a Machine Learning
model by considering the distances among the Automatic Weather Stations and their respective
daily temperature. The spatial distribution of daily temperature was estimated with the greatest
accuracy of 96% by Machine Learning models when compared with Kriging based approaches. A
continuous surface for daily temperature was created for all Geostatistical and Machine Learning
Models for the Uttarakhand State. Keywords: Spatial Interpolation, Geostatistical techniques,
Machine Learning techniques, Automatic Weather Stations (AWS)."

Role of machine learning techniques for spatial interpolation of environmental variables

"A problem that lends itself to the application of machine learning is classifying matched sources in the Galex (Galaxy Evolution Explorer) and SDSS (Sloan Digital Sky Survey) catalogs into stars and quasars based on color-color plots. The problem is daunting because stars and quasars are still inextricably mixed elsewhere in the color-color plots and no clear linear/non-linear boundary separates the two entities. Diversity and volume of samples add to the complexity of the problem. We explore the efficacy of neural network based classification techniques in discriminating between stars and quasars using GALEX and SDSS photometric data. Both sources have compact optical morphology but are very different in nature and are at very different distances. We have used those objects with associated spectroscopic information as our training-set and built neural network and ensemble classifiers that appropriately classify photometric samples without associated spectroscopic labels. Catalogs comprising of samples labelled using our classifiers can be further used in studies of photometric sources. The design of a novel Generative Adversarial Network (GAN) based classifier is proposed in the paper to tackle the classification problem. To evaluate the correctness of the classifiers, we report the accuracy and other performance metrics and find reasonably satisfactory range of 91-100 % . Keywords: Generative Adversarial Networks (GANs), Random Forests, SDSS Catalog, Virtual observatory tools 
"

Separating Stars from Quasars: Machine Learning Investigation Using Photometric Data

Improving Unsupervised Domain Adaptation with Variational Information Bottleneck

Semantics-Reinforced Networks for Question Generation

Keyphrases are multipurpose knowledge gems, rendering keyphrase extraction a very important document processing task. They constitute a concise summary of documents that is extremely useful both for human inspection and machine consumption, in support of a number of tasks in the field of Natural Language Processing, Machine Learning and Information Retrieval. To this regard, the aim of this tutorial is two-fold: (a) to provide an overview of the automatic keyphrase extraction task and (b) to familiarize participants with the keyphrase extraction process. Specifically, we will provide a well‐structured review of the existing work, offer interesting insights on the different evaluation approaches, highlight open issues such as the need for evaluation approaches that take the semantic similarity of predicted and golden keyphrases, present a comparative experimental study of popular techniques, and familiarize the audience with the keyphrase extraction process via a demo presentation using jupyter notebook, which will be available to the audience during the practical part of the tutorial. We expect the tutorial to help newcomers and veterans alike navigate the large amount of prior art and grasp its evolution.

Automatic Keyphrase Extraction from Text: A Walk-through

Many important AI problems involve an optimization component. Maximum satisfiability (MaxSAT) is an optimization version of Boolean satisfiability (SAT) that can be applied to a variety of computationally hard optimization problems that arise in AI. Constituting a declarative approach to optimization, modern state-of-the-art MaxSAT solvers allow for efficiently tackling NP-hard optimization problems arising from different applications via simple propositional encodings. This can be much less time-consuming than building a customized algorithmic solution specific to a particular optimization problem. Indeed, a well-engineered, generic MaxSAT solver can often be more efficient than a custom solution. Furthermore, MaxSAT offers an alternative to the more classical declarative optimization paradigm of integer programming (IP), especially in the many AI applications where the underlying problems are naturally represented with logical constraints. In this tutorial, we will introduce MaxSAT, give an overview of the different techniques that have been developed for solving MaxSAT, focusing on recent advances and covering both complete and incomplete MaxSAT solving, and explain via case studies how MaxSAT has been used to solve problems from different areas of AI. We will also outline major current research directions in MaxSAT. After this tutorial, you should be better equipped to apply modern MaxSAT technology in real-world settings.

Downloads

Next from ECAI 2020

A Deep Learning Approach to Space Weather Forecasting for Orbital Prediction

Similar lecture

GalaxAI: Machine Learning for Spacecraft Operations

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ECAI 2020

A Deep Learning Approach to Space Weather Forecasting for Orbital Prediction

Similar lecture

GalaxAI: Machine Learning for Spacecraft Operations

Downloads