Here Complemation is the development of technology that is designed to aid in complementing human skills and abilities with automation rather than replacing them –”complemation” (Schutte, 1999).
Complemation aims to enhance human cognition in part by using covet brain and body signals to prevent cognitive and attention lapses in Human Machine Teaming operations.
When humans are acting as passive monitors of autonomous systems, it is inherently difficult for them to fully understand what is going on due to lower levels of Cognitive Engagement. Here we propose various projects to investigate and monitor the features that influence the human cognitive processes involved in successful oversight, intervention, and teaming with automated systems.
At present, all that any vehicle knows of its rider it learns through the rider’s overt behavior. Vehicle accidents are primarily due to human error and human error is primarily due to human cognitive and attentional lapses. These lapses are detectable in covert brain and body signals before they are manifested in overt behavior.
The new complemation communication channels place emphasis on both the rider’s trust/comfort in an autonomous system and on and the rider’s ability to take control. When the rider has the ability to take control, trust models that assess the system’s trust in the rider will allow the system to conditionally respond to handoffs (unanticipated automation transitions) based on cognitive state determinations made immediately prior to a handoff.
Ethology allows us to pursue the development of control architectures that mimic the horse/rider interaction. We can apply this architecture to modern autonomous vehicle platforms. Then, like the horse and rider, vehicle becomes an effective member of the Human Robot Team. This reduces workload, improves situational awareness, and improves team efficiency allowing improved safety, trust, and neglect tolerance.
The H-Mode established a mode of autonomy that mimics the co-operative, symbiotic relationship between a horse and its rider. For example, Using Haptic Sensing, the rider is constantly aware of what the horse is doing, even while focusing the riders attention elsewhere. If the horse is unsure about where to go, it will slow down, and seek a new obstacle free path while trying to get the rider back into the loop. The horse might also be aware of how engaged the rider is and adjust its behavior. If a dangerous situation suddenly pops up, it will try to react before it is too late. The rider can let the horse choose its path without being out-of-the loop or take it on tight rein to reassert a more direct command.
These “modes” essentially map over into the control modes of autonomy and informs a unique control architecture.
The horse/rider model ot the H-Metaphor uses complemation research to establish communication channels between the robot and the human. This then is used to inform the vehicle of the “state” of its human team member by deploying off-the body remote sensing of brain and body signals (e.g., video, infrared and ultrasonic sensor technologies, face reading, smart-seat body position sensing, non-contact brain sensing, and unobtrusive haptic messaging.)
By studying the relationship between service animals such as horse and canines we can lean how to create efficient and effective human-machine teams.
Passenger/Occupant/Crew state monitoring when paired with complementation involves non-contact brain wave monitoring, thermal/visual cameras for driver, passenger states, interaction with vehicles and engagement, identifying key metrics for vehicle-passenger interaction, as well as key for deployment of Autonomous Vehicles. All of this, complemented with machine learning models developed to predict system performance from psychophysiological trust metrics create a cohesive experience.
Performance of the prediction models to be compared between models incorporating all trust metrics versus those incorporating only overt, off-body metrics to determine the effectiveness of a monitoring technology that employs only the unobtrusive metrics.
At its conclusion, the projects will combine knowledge to demonstrate a system which enables improved safety performance beyond a fully manual or fully autonomous system in a real-world setting.
Modern autonomous systems are, and will be in the future, dependent on the development of successful approaches to human–autonomy teaming. Here we define and explore “the system” consisting of the platform, the software and the human “operator”. Future automation should be designed to mitigate the risk associated with the autonomy conundrum by actively assessing and managing the humans’ engagement. This website introduces new approaches to automation that involve human cognition, human-animal teaming, and complemation concepts in automation and control.