Performance In Complex Environments Research Articles

Efficient 3D object perception and grasp planning for mobile manipulation in domestic environments

In this article, we describe efficient methods for tackling everyday mobile manipulation tasks that require object pick-up. In order to achieve real-time performance in complex environments, we focus our approach on fast yet robust solutions. For 3D perception of objects on planar surfaces, we develop scene segmentation methods that process depth images in real-time at high frame rates. We efficiently plan feasible, collision-free grasps for the segmented objects directly from the perceived point clouds to achieve fast execution times. We evaluate our approaches quantitatively in lab experiments and also report on the successful integration of our methods in public demonstrations at RoboCup@Home competitions in 2011 and 2012.

Efficient model for indoor radio paths computation

This paper presents a radio paths modeling framework for simulating RF coverage in complex indoor environments. We propose an algorithm which utilizes a geometric visibility graph of a building to traverse all possible bounded radio paths. These paths are needed for the computation of signal strength captured at a given receiver location. We have implemented the suggested algorithm and conducted a set of experiments to evaluate its performance in complex environments. The main conclusion is that the new algorithm is both (i) Accurate: predicts the signal strength inside complex buildings. (ii) Runtime efficient: requires only few seconds to compute all relevant radio paths, even when operating on complex structures containing thousands of walls.

Using After-Action Review Based on Automated Performance Assessment to Enhance Training Effectiveness

Training simulators have become increasingly popular tools for instructing humans on performance in complex environments. However, the question of how to provide individualized and scenario-specific assessment and feedback to students remains largely an open question. In this work, we follow-up on previous evaluations of the Automated Expert Modeling and Automated Student Evaluation (AEMASE) system, which automatically assesses student performance based on observed examples of good and bad performance in a given domain. The current study provides a rigorous empirical evaluation of the enhanced training effectiveness achievable with this technology. In particular, we found that students given feedback via the AEMASE-based debrief tool performed significantly better than students given only instructor feedback on two out of three domain-specific performance metrics.

PLATT: A flexible platform for experimental research on team performance in complex environments

The present article introduces PLATT, a recently developed task environment for controlled experimental research on team performance in complex environments. PLATT was developed to meet the research demands posed by the complexity that present-day teams face. It consists of a flexible, modular software architecture and research-specific scenarios. The scenarios can target various types of tasks (e.g., planning, problem solving, and decision making) in different operational contexts. Different software configurations can be used to investigate questions pertaining to team structure, team virtuality, and multiteam systems. We describe the software architecture, one of the scenarios, and the broad range of automated and embedded measurement possibilities that PLATT offers. To illustrate PLATT's possibilities, in the present article, we describe a number of experiments that have used PLATT for a variety of research questions. We conclude that PLATT meets the formulated research demands and provides researchers with a flexible platform to investigate the complex issues that present-day teams face.

A re-balanced scorecard: a strategic approach to enhance managerial performance in complex environments

This paper is a proposal to develop conceptual and practical frameworks for evolving corporations seeking to improve their managerial performance in complex environments with actionable strategies for dealing with social, environmental and corporate governance issues. These frameworks are coalesced by social contract theory that extends the traditional view of the firm as a nexus of contracts to a broader view of the firm as a nexus of social contracts. A re-balanced scorecard is proposed to induce and evaluate management performance that captures important dimensions and aspects of the frameworks established for firms strategically choosing to change their long term objectives to include those related to meeting their social contract obligations.

Systemic assessment of the effect of mental stress and strain on performance in a maritime ship-handling simulator

Two studies will be presented as part of a systemic approach for assessing performance in nautical simulators. A mixed-methods quasi-experimental field study (N=6) was conducted, aiming at discerning the systemic causes behind nautical students’ human errors during simulator exercises and to what extent these causes can be related to the layout of a new decision supporting display. Results indicate that all errors occur under the same kind of (demanding) man-machine interaction. Based on this, design requirements were proposed. The second study aims at exploring the impact of situation-related affective arousal on system safety. Anger- and frustration-like situations will be generated and the possible impact of these affect-laden situations on the risk- and error-related performance of simulator ship bridge crews assessed. The studies address different aspects of performance in complex environments and their modularity towards each other will be highlighted in the context of an overall systemic perspective.

Leader cognition: Improving leader performance through causal analysis

Prior qualitative research has indicated that leader performance in complex environments requires active analysis of the key causes operative in the organizational system. To provide some quantitative evidence to bear on this proposition, 160 undergraduates were presented with a computer simulation task that placed the participant in the role of a university president in charge of raising educational quality on campus. Training in the analysis of causes was provided, along with additional manipulations which varied the quality of case readings as well as levels of complexity in the environment. Overall, it was found that training in causal analysis significantly improved leader performance. The implications of these findings for understanding leader performance in complex environments are discussed.

Effects of field-of-view restriction on manoeuvring in a 3-D environment

Field-of-view (FOV) restrictions are known to affect human behaviour and to degrade performance for a range of different tasks. However, the relationship between human locomotion performance in complex environments and FOV size is currently not fully known. This paper examined the effects of FOV restrictions on the performance of participants manoeuvring through an obstacle course with horizontal and vertical barriers. All FOV restrictions tested (the horizontal FOV was either 30°, 75° or 120°, while the vertical FOV was always 48°) significantly reduced performance compared to the unrestricted condition. Both the time and the number of footsteps needed to traverse the entire obstacle course increased with a decreasing FOV size. The relationship between FOV restriction and manoeuvring performance that was determined can be used to formulate requirements for FOV restricting devices that are deployed to perform time-limited human locomotion tasks in complex structured environments, such as night-vision goggles and head-mounted displays used in training and entertainment systems.

Dribble files: Methodologies to evaluate learning and performance in complex environments

Research in the area of technology learning environments is tremendously complex. Tasks performed in these contexts are highly cognitive and mostly invisible to the observer. The nature of performance in these contexts is explained not only by the outcome but also by the process. However, evaluating the learning process with respect to tasks involving technology is difficult to measure objectively. As a result, audit trails (also called log or dribble files) are becoming extremely valuable tools in evaluating learning and performance in complex environments. This article reviews efforts of researchers in various areas, describes the nature of research using dribble files, and provides a framework on which investigators might build evaluations.

INCREMENTAL LEARNING OF PROCEDURAL PLANNING KNOWLEDGE IN CHALLENGING ENVIRONMENTS

Autonomous agents that learn about their environment can be divided into two broad classes. One class of existing learners, reinforcement learners, typically employ weak learning methods to directly modify an agent's execution knowledge. These systems are robust in dynamic and complex environments but generally do not support planning or the pursuit of multiple goals. In contrast, symbolic theory revision systems learn declarative planning knowledge that allows them to pursue multiple goals in large state spaces, but these approaches are generally only applicable to fully sensed, deterministic environments with no exogenous events. This research investigates the hypothesis that by limiting an agent to procedural access to symbolic planning knowledge, the agent can combine the powerful, knowledge-intensive learning performance of the theory revision systems with the robust performance in complex environments of the reinforcement learners. The system, IMPROV, uses an expressive knowledge representation so that it can learn complex actions that produce conditional or sequential effects over time. By developing learning methods that only require limited procedural access to the agent's knowledge, IMPROV's learning remains tractable as the agent's knowledge is scaled to large problems. IMPROV learns to correct operator precondition and effect knowledge in complex environments that include such properties as noise, multiple agents and time-critical tasks, and demonstrates a general learning method that can be easily strengthened through the addition of many different kinds of knowledge.

Roadmap-based motion planning in dynamic environments

In this paper, a new method is presented for motion planning in dynamic environments, that is, finding a trajectory for a robot in a scene consisting of both static and dynamic, moving obstacles. We propose a practical algorithm based on a roadmap that is created for the static part of the scene. On this roadmap, an approximately time-optimal trajectory from a start to a goal configuration is computed, such that the robot does not collide with any moving obstacle. The trajectory is found by performing a two-level search for a shortest path. On the local level, trajectories on single edges of the roadmap are found using a depth-first search on an implicit grid in state-time space. On the global level, these local trajectories are coordinated using an A/sup */-search to find a global trajectory to the goal configuration. The approach is applicable to any robot type in configuration spaces with any dimension, and the motions of the dynamic obstacles are unconstrained, as long as they are known beforehand. The approach has been implemented for both free-flying and articulated robots in three-dimensional workspaces, and it has been applied to multirobot motion planning, as well. Experiments show that the method achieves interactive performance in complex environments.

Themes Surrounding Novice Nurse Near-Miss and Adverse-Event Situations

The study purpose was to identify human performance factors that characterized novice nurse near-miss/adverse-event situations in acute-care settings. Increased focus on recruitment and retention of newly graduated registered nurses (RNs) in light of patient safety improvement goals will challenge healthcare educators and administrators. What we are beginning to learn about human performance issues during real work situations from patient safety research provides information related to human performance in complex environments that may guide education and system supports for novice RNs. Data collected during 8 retrospective interviews of novice RNs about details surrounding their individual near-miss or adverse event were analyzed for common themes. Nine themes were identified. Seven themes were present in at least 7 of the 8 cases and included environmental and social issues, as well as novice lack of expertise. Findings suggest that support for novice nurses in acute care environments requires attention to the following: consistent availability of expertise in light of workload unpredictability, the social climate regarding expectations of novice performers, realistic expectations of novice decision-making ability during complex situations even up to a year after graduation, and strategies to recognize and intervene when novices are at risk for error.

Application of Bayesian Network Modeling in a Training Support Tool

The instructor/trainer is a key decision making element in the training and assessment cycle designed to support performance in complex environments, There is an identified need for innovative tools and methods for instructors to use to evaluate performance in simulation based training environments. Bayesian network methods are being investigated as part of a project to provide integrated assessment, debrief and analysis support tools for instructors in an advanced tactics school.

Low-Reynolds number modelling of flows with transpiration

An improved low-Reynolds number model was adopted to predict the dynamic and thermal fields in flows with transpiration. The performance of the adopted model was first contrasted with the direct numerical simulation (DNS) data of channel flow with uniform wall injection and suction. The validity of the present model applied to flows with a high level of transpiration was further examined. To explore the model's performance in complex environments, the model was applied to simulate a transpired developing channel flow. By contrasting the predictions with DNS data and measurements, the results indicated that the present model reproduced correctly the deceleration and acceleration of the flow caused by the injection and suction from the permeable part of the wall. The turbulence structure of transpired flows was also well captured and the superior performance of the adopted model was reflected by the predicted correct level of ε with the maximum being located at both the injection and the suction walls. The predicted thermal field by the present model also compared favourably with the DNS data and measurements. Copyright © 2000 John Wiley & Sons, Ltd.

Effects of Leader Persistence and Environmental Complexity on Leadership Perceptions

The effects of managerial persistence with one course of action and environmental complexity on leadership perceptions were explored. Persistent leaders were evaluated more highly and attributed more leadership and responsibility than nonpersistent leaders, regardless of the level of environmental complexity. The results suggest that implicit beliefs about effective leader behavior may not vary across situations, thereby discouraging variety in leader behavior critical for performance in complex environments.

SOME RELATIONSHIPS BETWEEN CONCEPTUAL COMPLEXITY AND MENTAL ABILITIES

ABSTRACTSeveral studies have indicated that a measure of the complexity of the conceptual structure with which a person deals with stimuli has potential for predicting certain types of performance involving information processing. Many of the theoretical operations and characteristics of more conceptually complex modes of behavior resemble descriptions of the exercise of ability constructs. The main purpose of the present study was to explore the relationships between a measure of conceptual complexity—the Paragraph Completion Test (PCT)—and measures of several established abilities.Several hypotheses were advanced concerning the relatedness of conceptual complexity to verbal abilities, general reasoning abilities, divergent production abilities, mastery of logic, and a measure heavily saturated with Spearman's “g” factor.A battery of tests was administered to 189 high school students. The PCT was given twice during the testing sessions. Estimates of the reliability of the PCT were computed using an internal consistency method and the test‐retest procedure. Multiple regression analyses were performed using the ability tests as independent variables and the two administrations of the PCT as dependent variables. Two factor analyses were performed on the intercorrelation matrix of the ability tests and the conceptual complexity tests.The scores of the subjects were divided into two groups, designated high and low conceptual complexity. The interrelations of the abilities of the two groups of subjects were then compared by testing the hypothesis that high complexity subjects have a different organization of abilities as reflected in the ability test intercorrelations.The two methods of computing reliability both gave estimates of .24 for the first administration of the PCT. These estimates were probably too low inasmuch as the computed communality and squared multiple correlation of the PCT with the ability test battery were both larger.The results of the multiple regression analysis indicated that the best predictors of performance on the PCT at the first administration were tests of general reasoning and subtests of the Progressive Matrices Test. This finding was interpreted as suggesting that the general reasoning factor involves an ability to structure problems and to deal with complexity, where complexity is defined by the number of elements and relations in a problem. The best predictors of performance on the second administration of the PCT were the Simile Interpretation Test, the Utility Test, and the general reasoning tests.The factor analysis identified a factor marked by the PCT. This was interpreted as suggesting the existence of a capability of the subjects not being measured by the ability tests.It was suggested that some of the restrictions on measures of conceptual complexity are unnecessary. Among the indications found in this study was that multiple‐choice questions may be adequate for the assessment of complexity.It was recommended that further research in the area on conceptual complexity take more account of the role of ability constructs in determining performance in complex environments.