Automation of earth-moving industries (construction, mining and quarry) require automatic bucket-filling algorithms for efficient operation of front-end loaders. Autonomous bucket-filling is an open problem since three decades due to difficulties in developing useful earth models (soil, gravel and rock) for automatic control. Operators make use of vision, sound and vestibular feedback to perform the bucket-filling operation with high productivity and fuel efficiency. In this paper, field experiments with a small time-delayed neural network (TDNN) implemented in the bucket control-loop of a Volvo L180H front-end loader filling medium coarse gravel are presented. The total delay time parameter of the TDNN is found to be an important hyperparameter due to the variable delay present in the hydraulics of the wheel-loader. The TDNN network successfully performs the bucket-filling operation after an initial period (100 examples) of imitation learning from an expert operator. The demonstrated solution show only 26% longer bucket-filling time, an improvement over manual tele-operation performance.

Recent years have seen a surge of interest in using IoT systems for an increasingly diverse set of applications, with use cases ranging from medicine to mining. Due to the disparate needs of these applications, vendors are adopting a growing number of messaging protocols, encodings and semantics, which result in poor interoperability unless systems are explicitly designed to work together. Key efforts, such as Industry 4.0, put heavy emphasis on being able to compose arbitrary IoT systems to create emergent applications, which makes mitigating this barrier to interoperability a significant objective. In this paper, we present a theoretical method for translating message payloads in transit between endpoints, complementing previous work on protocol translation. The method involves representing and analyzing encoding syntaxes with the aim of identifying the concrete translations that can be performed without risk of syntactic data loss. While the method does not facilitate translation between all possible encodings or semantics, we believe that it could be extended to enable such translation.

Many use cases coming out of initiatives such as Industry 4.0 and Ubiquitous Computing require that systems be able to cooperate by negotiating about and agreeing on the exchange of arbitrary values. While solutions able to facilitate such negotiation exist, they tend to either be domain-specific or lack mechanisms for non-repudiation, which make them unfit for the heterogeneity and scale of many compelling applications. In this paper, we present the Exchange Network, a general-purpose and implementation-independent architecture for digital negotiation and non-repudiable exchanges of tokens, which are symbolic representations of arbitrary values. We consider the implications of implementing the architecture in three different ways, using a common database, a blockchain, and our own Signature Chain data structure, which we also describe. We demonstrate the feasibility of the architecture by outlining our own implementation of it and also describe a supply-chain scenario inspired by one transportation process at Volvo Trucks.

Remote operation is a step toward the automation of mobile working machines. Safe and efficient teleremote operation requires good-quality video feedback. Varying radio conditions make it desirable to adapt the video sending rate of cameras to make the best use of the wireless capacity. The adaptation should be able to prioritize camera feeds in different directions depending on motion, ongoing tasks, and safety concerns. Self-Clocked Rate Adaptation for Multimedia (SCReAM) provides a rate adaptation algorithm for these needs. SCReAM can control the compression used for multiple video streams using differentiating priorities and thereby provide sufficient congestion control to achieve both low latency and high video throughput. We present results from the testing of prioritized adaptation of four video streams with SCReAM over LTE and discuss how such adaptation can be useful for the teleremote operation of working machines.

This paper presents experimental results with tele-remote operation of a wheel-loader and proposes a method to semi-automate the process. The different components of the tele-remote setup are described in the paper. We focus on the short loading cycle, which is commonly used at quarry and construction sites for moving gravel from piles onto trucks. We present results from short-loading-cycle experiments with three operators, comparing productivity between tele-remote operation and manual operation. A productivity loss of 42% with tele-remote operation motivates the case for more automation. We propose a method to automate the bucket-filling process, which is one of the key operations performed by a wheel-loader.

Automatic bucket filling is an open problem since three decades. In this paper, we address this problem with supervised machine learning using data collected from manual operation. The range-normalized actuations of lift joystick, tilt joystick and throttle pedal are predicted using information from sensors on the machine and the prediction errors are quantified. We apply linear regression, k-nearest neighbors, neural networks, regression trees and ensemble methods and find that an ensemble of neural networks results in the most accurate predictions. The prediction root-mean-square-error (RMSE) of the lift action exceeds that of the tilt and throttle actions, and we obtain an RMSE below 0.2 for complete bucket fillings after training with as little as 135 bucket filling examples

Distributed ledger technologies, such as blockchain systems, have in recent years emerged as promising platforms for machine-to-machine commerce and other forms of multi-stakeholder applications. However, despite the potential demonstrated by projects such as Bitcoin, Ethereum, and Hyperledger Fabric, the disk space typically required to host a copy of a ledger may be prohibitively large for many categories of devices. In this paper, we introduce an approach for reducing ledger size in blockchain systems, based on arbitrary pruning predicate functions, allowing each network participant to independently select and remove any already applied transactions. We also show that if only pruning certain ledger transactions, the ability to derive an unmodified state data structure from the remaining transactions is maintained. The approach is validated through a supply chain use case utilizing a modified version of Hyperledger Fabric, in which ledger size is reduced by about 84.49% via selective transaction pruning.

Driver awareness of current winter road conditions (RCs) is known to affect the frequency of accidents due to sudden changes in these conditions. For example, partially icy roads that appear during autumn in northern areas typically result in collisions and ditch runs unless the drivers are generally aware of the situation. Availing motorists who drive under winter RCs of enhanced information is therefore highly desirable to increase their awareness of hazardous driving conditions. Such conditions need to be predicted ahead of time and presented to drivers before they attempt slippery road sections. Moreover, the identification of slippery RCs should quickly trigger targeted road maintenance to reduce the risk of accidents. This study presents a scalable and reusable collaborative intelligent transport system, herein referred to as an RC information system (RCIS). RCIS provides accurate RC predictions and forecasts based on RC measurements, road weather observations, and short-term weather forecasts. The prediction methods in the context of the distributed RCIS have been tested using a prototype implementation. These tests confirmed that these inputs could be combined into useful and accurate information about winter RCs that can be adapted for different types of users.

A wheel loader is an earth-moving machine used in construction sites, gravel pits and mining to move blasted rock, soil and gravel. In the presence of a nearby dump truck, the wheel loader is said to be operating in a short loading cycle. This paper concerns the moving of material (soil, gravel and fragmented rock) by a wheel loader in a short loading cycle with more emphasis on the loading step. Due to the complexity of bucket-environment interactions, even three decades of research efforts towards automation of the bucket loading operation have not yet resulted in any fully autonomous system. This paper highlights the key challenges in automation and tele-remote operation of earth-moving machines and provides a survey of different areas of research within the scope of the earth-moving operation. The survey of publications presented in this paper is conducted with an aim to highlight the previous and ongoing research work in this field with an effort to strike a balance between recent and older publications. Another goal of the survey is to identify the research areas in which knowledge essential to automate the earth moving process is lagging behind. The paper concludes by identifying the knowledge gaps to give direction to future research in this field.

The automation of bucket loading for repetitive tasks of earth-moving operations is desired in several applications at mining sites, quarries and construction sites where larger amounts of gravel and fragmented rock are to be moved. In load and carry cycles the average bucket weight is the dominating performance parameter, while fuel efficiency and loading time also come into play with short loading cycles. This paper presents the analysis of data recorded during loading of different types of gravel piles with a Volvo L110G wheel loader. Regression models of lift and tilt actions are fitted to the behavior of an expert driver for a gravel pile. We present linear regression models for lift and tilt action that explain most of the variance in the recorded data and outline a learning approach for solving the automatic bucket loading problem. A general solution should provide good performance in terms of average bucket weight, cycle time of loading and fuel efficiency for different types of material and pile geometries. We propose that a reinforcement learning approach can be used to further refine models fitted to the behavior of expert drivers, and we briefly discuss the scooping problem in terms of a Markov decision process and possible value functions and policy iteration schemes.