Error modeling and calibration of exteroceptive sensors for accurate mapping applications

Reliable robotic perception and planning are critical to performing autonomous actions in uncertain, unstructured environments. In field robotic systems, automation is achieved by interpreting exteroceptive sensor information to infer something about the world. This is then mapped to provide a consistent spatial context, so that actions can be planned around the predicted future interaction of the robot and the world. The whole system is as reliable as the weakest link in this chain. In this paper, the term mapping is used broadly to describe the transformation of range-based exteroceptive sensor data (such as LIDAR or stereo vision) to a fixed navigation frame, so that it can be used to form an internal representation of the environment. The coordinate transformation from the sensor frame to the navigation frame is analyzed to produce a spatial error model that captures the dominant geometric and temporal sources of mapping error. This allows the mapping accuracy to be calculated at run time. A generic extrinsic calibration method for exteroceptive range-based sensors is then presented to determine the sensor location and orientation. This allows systematic errors in individual sensors to be minimized, and when multiple sensors are used, it minimizes the systematic contradiction between them to enable reliable multisensor data fusion. The mathematical derivations at the core of this model are not particularly novel or complicated, but the rigorous analysis and application to field robotics seems to be largely absent from the literature to date. The techniques in this paper are simple to implement, and they offer a significant improvement to the accuracy, precision, and integrity of mapped information. Consequently, they should be employed whenever maps are formed from range-based exteroceptive sensor data. © 2009 Wiley Periodicals, Inc.

Combining recommender and reputation systems to produce better online advice

Although recommender systems and reputation systems have quite different theoretical and technical bases, both types of systems have the purpose of providing advice for decision making in e-commerce and online service environments. The similarity in purpose makes it natural to integrate both types of systems in order to produce better online advice, but their difference in theory and implementation makes the integration challenging. In this paper, we propose to use mappings to subjective opinions from values produced by recommender systems as well as from scores produced by reputation systems, and to combine the resulting opinions within the framework of subjective logic.

Wetlands : a perspective for mapping, monitoring, and modeling

Experience gained from numerous projects conducted by the U.S. Environmental Protection Agency's (EPA) Environmental Monitoring Systems Laboratory in Las Vegas, Nevada has provided insight to functional issues of mapping, monitoring, and modeling of wetland habitats. Three case studies in poster form describe these issues pertinent to managing wetland resources as mandated under Federal laws. A multiphase project was initiated by the EPA Alaska operations office to provide detailed wetland mapping of arctic plant communities in an area under petroleum development pressure. Existing classification systems did not meet EPA needs. Therefore a Habitat Classification System (HCS) derived from aerial photography was compiled. In conjunction with this photointerpretive keys were developed. These products enable EPA personnel to map large inaccessible areas of the arctic coastal plain and evaluate the sensitivity of various wetland habitats relative to petroleum development needs.

A Lagrangian approach to nonlinear modeling of anti-roll tanks

This paper presents two novel nonlinear models of u-shaped anti-roll tanks for ships, and their linearizations. In addition, a third simplified nonlinear model is presented. The models are derived using Lagrangian mechanics. This formulation not only simplifies the modeling process, but also allows one to obtain models that satisfy energy-related physical properties. The proposed nonlinear models and their linearizations are validated using model-scale experimental data. Unlike other models in the literature, the nonlinear models in this paper are valid for large roll amplitudes. Even at moderate roll angles, the nonlinear models have three orders of magnitude lower mean square error relative to experimental data than the linear models.

Hypergraph-based modeling of ad-hoc business processes

Process models are usually depicted as directed graphs, with nodes representing activities and directed edges control flow. While structured processes with pre-defined control flow have been studied in detail, flexible processes including ad-hoc activities need further investigation. This paper presents flexible process graph, a novel approach to model processes in the context of dynamic environment and adaptive process participants’ behavior. The approach allows defining execution constraints, which are more restrictive than traditional ad-hoc processes and less restrictive than traditional control flow, thereby balancing structured control flow with unstructured ad-hoc activities. Flexible process graph focuses on what can be done to perform a process. Process participants’ routing decisions are based on the current process state. As a formal grounding, the approach uses hypergraphs, where each edge can associate any number of nodes. Hypergraphs are used to define execution semantics of processes formally. We provide a process scenario to motivate and illustrate the approach.

The Interaction of lean and building information modeling in construction

Lean construction and building information modeling (BIM) are quite different initiatives, but both are having profound impacts on the construction industry. A rigorous analysis of the myriad specific interactions between them indicates that a synergy exists which, if properly understood in theoretical terms, can be exploited to improve construction processes beyond the degree to which it might be improved by application of either of these paradigms independently. Using a matrix that juxtaposes BIM functionalities with prescriptive lean construction principles, 56 interactions have been identified, all but four of which represent constructive interaction. Although evidence for the majority of these has been found, the matrix is not considered complete but rather a framework for research to explore the degree of validity of the interactions. Construction executives, managers, designers, and developers of information technology systems for construction can also benefit from the framework as an aid to recognizing the potential synergies when planning their lean and BIM adoption strategies.

Near knowledge : inductive learning systems in law

Induction is an interesting model of legal reasoning, since it provides a method of capturing initial states of legal principles and rules, and adjusting these principles and rules over time as the law changes. In this article I explain how Artificial Intelligence-based inductive learning algorithms work, and show how they have been used in law to model legal domains. I identify some problems with implementations undertaken in law to date, and create a taxonomy of appropriate cases to use in legal inductive inferencing systems. I suggest that inductive learning algorithms have potential in modeling law, but that the artificial intelligence implementations to date are problematic. I argue that induction should be further investigated, since it has the potential to be an extremely useful mechanism for understanding legal domains.

Built heritage management systems : the framework of a digital tool for the conservation of Brisbane City Hall

Brisbane City Hall (BCH) is arguably one of Brisbane’s most notable and iconic buildings. Serving as the public’s central civic and municipal building since 1930, the importance of this heritage listed building to cultural significance and identity is unquestionable. This attribute is reflected within the local government, with a simplified image of the halls main portico entrance supplying Brisbane City Council with its insignia and trademark signifier. Regardless of these qualities, this building has been neglected in a number of ways, primarily in the physical sense with built materials, but also, and just as importantly, through inaccurate and undocumented works. Numerous restoration and renovation works have been undertaken throughout BCH’s lifetime, however the records of these amendments are far and few between. Between 2010 and 2013, BCH underwent major restoration works, the largest production project undertaken on the building since its initial construction. Just prior to this conservation process, the full extent of the buildings deterioration was identified, much of which there was little to no original documentation of. This has led to a number of issues pertaining to what investigators expected to find within the building, versus what was uncovered (the unexpected), which have resulted directly from this lack of data. This absence of record keeping is the key factor that has contributed to the decay and unknown deficiencies that had amassed within BCH. Accordingly, this raises a debate about the methods of record keeping, and the need for a more advanced process that is able to be integrated within architectural and engineering programs, whilst still maintaining the ability to act as a standalone database. The immediate objective of this research is to investigate the restoration process of BCH, with focus on the auditorium, to evaluate possible strategies to record and manage data connected to building pathology so that a framework can be developed for a digital heritage management system. The framework produced for this digital tool will enable dynamic uses of a centralised database and aims to reduce the significant data loss. Following an in-depth analysis of this framework, it can be concluded that the implementation of the suggested digital tool would directly benefit BCH, and could ultimately be incorporated into a number of heritage related built form.

Non-formal techniques for requirements elicitation, modeling, and early assessment for services

Designing systems for multiple stakeholders requires frequent collaboration with multiple stakeholders from the start. In many cases at least some stakeholders lack a professional habit of formal modeling. We report observations from two case studies of stakeholder-involvement in early design where non-formal techniques supported strong collaboration resulting in deep understanding of requirements and of the feasibility of solutions.

Interaction modeling using BPMN

Process choreographies describe interactions between different business partners and the dependencies between these interactions. While different proposals were made for capturing choreographies at an implementation level, it remains unclear how choreographies should be described on a conceptual level.While the Business Process Modeling Notation (BPMN) is already in use for describing choreographies in terms of interconnected interface behavior models, this paper will introduce interaction modeling using BPMN. Such interaction models do not suffer from incompatibility issues and are better suited for human modelers. BPMN extensions are proposed and a mapping from interaction models to interface behavior models is presented.

Facilitating query decomposition in query language modeling by association rule mining using multiple sliding windows

This paper presents a novel framework to further advance the recent trend of using query decomposition and high-order term relationships in query language modeling, which takes into account terms implicitly associated with different subsets of query terms. Existing approaches, most remarkably the language model based on the Information Flow method are however unable to capture multiple levels of associations and also suffer from a high computational overhead. In this paper, we propose to compute association rules from pseudo feedback documents that are segmented into variable length chunks via multiple sliding windows of different sizes. Extensive experiments have been conducted on various TREC collections and our approach significantly outperforms a baseline Query Likelihood language model, the Relevance Model and the Information Flow model.

Modeling nitrous oxide emissions from irrigated agriculture : testing DayCent with high-frequency measurements

A unique high temporal frequency dataset from an irrigated cotton-wheat rotation was used to test the agroecosystem model DayCent to simulate daily N2O emissions from sub-tropical vertisols under different irrigation intensities. DayCent was able to simulate the effect of different irrigation intensities on N2O fluxes and yield, although it tended to overestimate seasonal fluxes during the cotton season. DayCent accurately predicted soil moisture dynamics and the timing and magnitude of high fluxes associated with fertilizer additions and irrigation events. At the daily scale we found a good correlation of predicted vs. measured N2O fluxes (r2 = 0.52), confirming that DayCent can be used to test agricultural practices for mitigating N2O emission from irrigated cropping systems. A 25 year scenario analysis indicated that N2O losses from irrigated cotton-wheat rotations on black vertisols in Australia can be substantially reduced by an optimized fertilizer and irrigation management system (i.e. frequent irrigation, avoidance of excessive fertiliser application), while sustaining maximum yield potentials.

Tensor-based item recommendation using probabilistic ranking in social tagging systems

A common problem with the use of tensor modeling in generating quality recommendations for large datasets is scalability. In this paper, we propose the Tensor-based Recommendation using Probabilistic Ranking method that generates the reconstructed tensor using block-striped parallel matrix multiplication and then probabilistically calculates the preferences of user to rank the recommended items. Empirical analysis on two real-world datasets shows that the proposed method is scalable for large tensor datasets and is able to outperform the benchmarking methods in terms of accuracy.

Modeling and analysis of a novel variable-speed cage induction generator

This paper introduces a novel cage induction generator and presents a mathematical model, through which its behavior can be accurately predicted. The proposed generator system employs a three-phase cage induction machine and generates single-phase and constant-frequency electricity at varying rotor speeds without an intermediate inverter stage. The technique uses any one of the three stator phases of the machine as the excitation winding and the remaining two phases, which are connected in series, as the power winding. The two-series-connected-and-one-isolated (TSCAOI) phase winding configuration magnetically decouples the two sets of windings, enabling independent control. Electricity is generated through the power winding at both sub- and super-synchronous speeds with appropriate excitation to the isolated single winding at any frequency of generation. A dynamic mathematical model, which accurately predicts the behavior of the proposed generator, is also presented and implemented in MATLAB/Simulink. Experimental results of a 2-kW prototype generator under various operating conditions are presented, together with theoretical results, to demonstrate the viability of the TSCAOI power generation. The proposed generator is simple and capable of both storage and retrieval of energy through its excitation winding and is expected to be suitable for applications, such as small wind turbines and microhydro systems.

Degradation modeling and monitoring of machines using operation-specific hidden Markov models

In this paper, a novel data-driven approach to monitoring of systems operating under variable operating conditions is described. The method is based on characterizing the degradation process via a set of operation-specific hidden Markov models (HMMs), whose hidden states represent the unobservable degradation states of the monitored system while its observable symbols represent the sensor readings. Using the HMM framework, modeling, identification and monitoring methods are detailed that allow one to identify a HMM of degradation for each operation from mixed-operation data and perform operation-specific monitoring of the system. Using a large data set provided by a major manufacturer, the new methods are applied to a semiconductor manufacturing process running multiple operations in a production environment.