SUMMARYIn this paper we present a cognitive architecture inspired on the biological functioning of the motor system in humans. To test the model, we built a robotic hand with a Lego Mindstorms TM kit. Then, through communication between the architecture and the robotic hand, the latter was able to perform the movement of the fingers, which therefore allowed it to perform grasping of some objects. In order to obtain these results, the architecture performed a conversion of the activation of motor neuron pools into specific degrees of servo motor movement. In this case, servo motors acted as muscles, and degrees of movement as exerted muscle force. Finally, this architecture will be integrated with high-order cognitive functions towards getting automatic motor commands generation, through planning and decision making mechanisms. key words: cognitive architecture, artificial intelligence, robotic hand
In humans, the vestibular system along with other sensory and motor systems is responsible for three cognitive functions that support mobility. First, is responsible for the balance of the body. Second, it allows humans to maintain the head stabilized. Finally, whenever the body or head are in motion, it maintains the visual gaze on a desired target. These tasks are performed using an array of sensors that are located within the inner ear. This paper describes the design and implementation of a synthetic model of the human vestibular system. The model is based on neurophysiological evidence, which makes it necessary to model all of the neural and physical components involved in the balance of the body. The model includes a component for each of the sensors, cortical and subcortical neural structures. It also defines and generates the necessary motor output signals. The proposed model was connected to a Bioloid® Premium humanoid robot to simulate the motor output and the proprioceptive inputs. The physical tests resulted inconclusive due to the fact that the controller on the robot was incapable of handling the necessary information for the tests. However, even though the results were not the desired, the communication between the sensors and the architecture, as well as the processing inside the architecture satisfied all of the authors' expectations.
One of the biggest challenges in the development of virtual creatures is provide them with mechanisms that allow them to generate behaviors autonomously. Therefore we think it is possible to use agent architectures to endow those virtual entities, so to speak, with architectures that can serve as the brain for these entities. In this paper we present an agent architecture that takes as inspiration for its creation, biological and psychological theories that consider that needs are the cause of behaviors in living beings. To test the model, a predator-prey scenario was selected. In this virtual world, the creatures do not have an established purpose; therefore their survival completely depends on the satisfaction of their needs. As a result of this work we obtained autonomous entities that can survive in a predator-prey environment.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.