Abstract Software‐based expert systems typically use an iterative algorithm for rule‐fact network processing. However, other approaches to expert system implementation also exist. This paper proposes simplified quantum principle expert systems and compares the computational efficiency and comparative efficacy of four types of expert systems: software‐based Boolean expert systems, electronic (gate‐based) expert systems, software‐based fractional value (SBFV) expert systems and simplified quantum principal expert systems (SQPES). For two expert system types, software‐based Boolean and SBFV, back chaining and mixed changing techniques, respectively, are also evaluated. The performance of these systems approaches is compared quantitatively, and the efficacy of the different systems types are compared both qualitatively and quantitatively.

When a cyberattack occurs, tracking the attack back to an actual individual person can be problematic. Even identifying the workstation or device that it originated from does not necessarily identify the attacker, as the attacking device could, itself, be compromised. A system to determine whether activities that occur are from the same user or not facilitates forensic analysis as well as the detection of concurrent attacks from different devices by the same user. This paper proposes a system for identifying attackers based on behaviors expressed via their use of the Bash command line interface, the most common shell on Linux distributions. Prior systems were limited by issues such as requiring labeled user data which is difficult to acquire or not being specific enough to monitor individual persons. The approach proposed herein does not require labeled data and is specific enough to target individual users. The proposed system analyzes the level of variance between commands used and calculates an anomaly score for each given command. It uses these anomaly scores to compare Bash history sets together to identify if they were created by the same user.

There are many factors that play a part in how a student performs during a course. Some of these factors can be looked at in hindsight to help suggest why a student may or may not have done well in class. This poster assesses the potential of predicting student performance in an introductory computer science class using information about students' preparation, attitudes and study habits. An expert system has been utilized for this purpose. The expert system accepts data related to seven different categories of preparation, belief and attitude and, through the partial activation of multiple rules, predicts an outcome for each student on the post-test (which should correlate with and is used as a surrogate for the student's final course grade). This poster presents our findings and correlation between certain factors relevant to students' success. It shows how these factors can be used to predict a student's grade and discusses the significance of being able to do so. Thus far, the results from the first sample of data appear promising. The work has shown that certain groups of self-assessment questions have clearly out performed other groups in predicting final test grades. The use of expert system techniques to predict student grades can provide an insight on different factors that affect student performance. Knowledge of characteristics that may lead to poor performance allows instructors to potentially identify students who may be at risk of a low grade and need supplemental instruction, tutoring or other support at the outset of the semester.

Abstract The blackboard architecture has particular utility in applications where conclusions must be drawn in the absence of human involvement. Through properly crafted rules, this approach can determine what a collection of data may mean. This can be utilized by other onboard software to limit transmission to only relevant data or conclusions. Some applications, such as robotic exploration, may preclude or severely limit the availability of controller‐to‐craft‐agent communications. In this case, the maintenance of the system must be performed autonomously.This paper discusses common maintenance tasks that may be relevant to a long‐running blackboard‐based system (e.g. clearing extraneous data from the blackboard regularly to improve performance, archiving relevant but infrequently used data for performance enhancement). A system for automating these tasks is presented that is suitable for most blackboard‐style systems. Its particular utility to those that cannot be easily accessed is discussed. Quantitative analysis of the value of removing rules from the system is performed and presented, and these results are extrapolated to several prospective blackboard maintenance approaches.

Cryptography in the domain of small satellites is a relatively new area of research. Compared to typical desktop computers, small satellites have limited bandwidth, processing power, and battery power. Many of the current encryption schemes were developed for desktop computers and servers, and as such may be unsuitable for small satellites. In addition, most cryptographic research in the domain of small satellites focuses on hardware solutions, which can be problematic given the limited space requirements of small satellites. This paper investigates potential software solutions that could be used to encrypt and decrypt data on small satellites and other devices with similarly limited resources. Specifically, this paper presents an implementation of an encryption algorithm based on chaos theory and compares and contrasts it with existing cryptographic schemes used in desktop computing and lightweight applications.

With internet-connected, SCADA and cyber-physical systems becoming the next battlefield for crime and warfare, technologies for defending and attacking these systems are growing in prevalence. For entities with significant asset collections that are prospectively vulnerable to this type of an attack, autonomous response, retaliation and attack capabilities are necessary to respond to a growing threat from numerous sectors. This paper presents a command and control technique for cyberwarfare based on the Blackboard Architecture. It discusses the utility of this approach and proposes a distributed command system that can run across multiple nodes of various types.

This article presents an approach for decision-making in support of the control of an autonomous system of multiple tiers of robots (e.g., satellite, aerial and ground) based on the Blackboard architectural style. Under the proposed approach, the system evaluates prospective approaches for goal satisfaction (identified by user selected final rules), identifies the lowest-cost solution and determines the best path to achieving the goal, via the analysis of the Blackboard rule and action set. Two different approaches to this rule and action path generation are discussed. This article presents the proposed Blackboard-style architecture for autonomous multi-tier control and describes its implementation. The benefits and drawbacks of the Blackboard-style approach are analysed, its extrapolation to the control of multiple heterogeneous craft is presented and the tradeoffs between the two approaches to rule-path generation are assessed.

State-of-the-art visual odometry algorithms achieve remarkable efficiency and accuracy. Under realistic conditions, however, tracking failures are inevitable and to continue tracking, a recovery strategy is required. In this paper, we propose a relocalization system that enables realtime, 6D pose recovery for wide baselines. Our approach targets specifically resource-constrained hardware such as mobile phones. By exploiting the properties of low-complexity binary feature descriptors, nearest-neighbor search is performed efficiently using Locality Sensitive Hashing. Our method does not require time-consuming offline training of hash tables and it can be applied to any visual odometry system. We provide a thorough evaluation of effectiveness, robustness and runtime on an indoor test sequence with available ground truth poses. We investigate the system parameterization and compare the relocalization performance for the three binary descriptors BRIEF, unscaled BRIEF and ORB. In contrast to previous work on mobile visual odometry, we are able to quickly recover from tracking failures within maps with thousands of 3D feature points.

This paper presents a simplified approach to performing the Intelligent Water Drops (IWD) process. This approach is designed to be comparatively lightweight while approximating the results of the full IWD process. The Simplified Intelligent Water Drops (SIWD) approach is specifically designed for applications where IWD must be run in a computationally limited environment (such as on a robot, UAV or small spacecraft) or where performance speed must be maximized for time sensitive applications. The SWID approach is described and compared and contracted to the base IWD approach.