China
Africa
Explore chapters and articles related to this topic
Human Performance Challenges for the Future Force: Lessons from Patriot after the Second Gulf War
Published in Pamela Savage-Knepshield, John Martin, John Lockett, Laurel Allender, Designing Soldier Systems, 2018
The Catch 22 sequence illustrates the logical inconsistency in the way that many current systems are designed and later used. To begin, automation is justified because humans are becoming the weakest link in increasingly complex defense systems. But humans are left on the control loop to verify that the machine’s performance is in accord with human intent. However, in many situations—including Patriot during OIF—this can be an unrealistic task demand. Humans make very poor monitors of complex processes (Davies and Parasuraman 1982). Moreover, the search for the disconfirming evidence necessary to override a machine decision or recommendation can require more resources (time, knowledge, and competence) than controllers can bring to bear on the situation. Norman (2007) states bluntly that any system operating between the extremes of full automation (no human involvement) and complete manual control (no machine involvement in at least the cognitive aspects of process or task control) occupies a “dangerous middle ground.” In a similar vein, Woods and Hollnagel (2006: 125) cite what is termed “Robin Murphy’s Law” to highlight the problem of being on this dangerous middle ground. The gist of Robin Murphy’s Law is that any automated system will fall short of its anticipated level of autonomy, and this shortfall will lead to problems in achieving and maintaining effective human oversight. Maintaining effective human supervisory control of a complex process or task setting is a problematic issue.
Automation and Human Performance
Published in Christopher D. Wickens, Justin G. Hollands, Simon. Banbury, Raja. Parasuraman, Engineering Psychology and Human Performance, 2015
Christopher D. Wickens, Justin G. Hollands, Simon. Banbury, Raja. Parasuraman
Analyses of automation-related incidents and accidents reveal that the functionality of the automation has a major influence on how well human operators interact with automation in meeting their system performance goals. The different functions that automation can take have been described in a number of ways. Automation is not all or none, but can vary across a continuum of levels, from the lowest level of fully manual performance (no automation) to the highest level of full automation. Sheridan and Verplanck (1978), in proposing the concept of supervisory control, first suggested a taxonomy of 10 such levels of automation. Supervisory control refers to a system in which a human operator does not directly operate on the physical plant being controlled but does so through an intermediary, usually a computer, that has effectors to act on the environment based on information obtained from sensors (Sheridan, 2002; Sheridan & Parasuraman, 2006).
1
Published in Marilyn Sue Bogner, Human Error in Medicine, 2018
Supervisory control also involves the optimum planning of actions and the scheduling of their efficient execution. At any given time there are multiple tasks, each of which is intrinsically appropriate, yet they cannot all be done at once. Each action must be interleaved with the myriad other concurrent activities. The expert anesthetist considers many factors in planning and adapting optimum action sequences, including: Preconditions necessary for carrying out the actions (e.g., it is impossible to measure the blood flow generated by the heart if the appropriate catheter needed to make the measurement is not already in place inside the heart).Constraints on the proposed actions (e.g., it is impossible to check the size of the patient’s pupils when the head is fully draped in the surgical field).Side effects of the proposed actions.Rapidity and ease of implementing the actions.Certainty of success of the actions.Reversibility of the action and the “cost of being wrong.”Cost of the action in terms of attention and of resources.
Distributed adaptive fault-tolerant supervisory control for leader-following systems with actuator faults
Published in International Journal of Systems Science, 2022
Jianye Gong, Bin Jiang, Yajie Ma, Xiaodong Han, Jianglei Gong
In practice, actuators may often suffer from certain faults, especially in the complicated networked communication systems, and these faults may result in the system performance degradation or even system states instability. In order to guarantee the system's security, fault-tolerant control (FTC) becomes an effective tool and has attracted extensive attention (e.g. Hua et al., 2018; Jiang et al., 2006; C. Liu et al., 2019, 2018; X. Wang & Yang, 2020; Yang et al., 2019; Zou et al., 2020). Fruitful results on the cooperative FTC scheme have been obtained. In Zou et al. (2020), the distributed fault-tolerant consensus tracking control problem was investigated for a class of nonlinear multiagent systems with heterogeneous and switched agent's dynamics. In C. Liu et al. (2019), a novel sliding-mode FTC scheme was designed for heterogeneous multiagent systems with actuator faults and disturbances. Cooperative fault-tolerant control protocols for multiagent systems subject to both actuator bias and loss of effectiveness faults were proposed in Hua et al. (2018) and X. Wang and Yang (2020). Despite these efforts, how to design the control scheme to ensure the stability of system states and improve the system performance by combining the FTC technique and the supervisory control strategy is important. To the best of our knowledge, how to apply supervisory control methods to this field is still a challenging and open subject.
How selfish individuals achieve unselfish goals: majority-based progressive control of discrete event systems
Published in International Journal of Control, 2020
Supervisory control is a feedback control for discrete event systems (DESs) of which state transitions happen by irregularly occurring events (Ramadge & Wonham, 1987). It provides systematic ways to achieve desirable behaviours by enabling or disabling controllable events following the observation of system behaviours. Supervisory control theory can be applied to many dynamical systems including power systems (Afzalian, Niaki, Iravani, & Wonham, 2009), biological systems (Baldissera, Cury, & Raisch, 2016), manufacturing systems (Hu & Zhou, 2015), economic systems (Park, 2011), and software systems (Phoha, Nadgar, Ray, & Phoha, 2004). In this paper, we present a new supervisory control framework called majority-based progressive control inspired by a principle of democratic progress, namely, making more people meet their interests or rights. For this purpose, we suppose that local supervisors have their own private specifications as well as an additional global specification which the overall system must achieve. The aim of progressive control is to increase the number of local supervisors meeting private specifications as the controlled system undergoes transitions. Here, the progressive behaviour means that the number of local supervisors meeting private specifications is not strictly increasing, but non-decreasing with at least one transition at which the number increases. This notion is targeted to imitate the gradual development of democracy.
SWM and urban water: Smart management for an absurd system?
Published in Water International, 2020
Currently, there are options for each engineered water services system to have dedicated, and in many cases advanced and elegant, automated and semi-automated management systems. Supervisory control and data acquisition systems, for example, provide essential information and support to ensure that potable water and wastewater systems operate reliably, while using energy and chemicals efficiently. This is fine as far as it goes, but the urban water complex also includes, and in fact relies on, the natural streams, creeks, rivers and lakes flowing through each urban area. These features are more than just drainage for stormwater or receiving waters for wastewater effluents. They provide one of the most important functions for successful stormwater and wastewater management in that they receive and convey the outputs of engineered urban water systems. For potable water systems, surface waters and aquifers clearly provide the most important service: water itself.