Iñaki Rañó
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Iñaki Rañó has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Augusto Gómez Eguíluz
GRVC Robotics Lab Sevilla, Universidad de Sevilla, Seville, Spain
8 shared publications - Sonya Coleman
Intelligent Systems Research Centre, University of Ulster, Derry, UK
6 shared publications - Thomas Martin McGinnity
School of Science and Technology, Nottingham Trent University, Nottingham, UK
6 shared publications - KongFatt Wong-Lin
Intelligent Systems Research Centre, School of Computing, Engineering and Intelligent Systems, Ulster University, Magee campus, Derry∼Londonderry, Northern Ireland, UK
4 shared publications - Mehdi Khamassi
Institute of Intelligent Systems and Robotics, CNRS, Sorbonne Université, Paris, France
4 shared publications - Kathryn Merrick
School of Engineering and Information Technology, University of New South Wales, Canberra, ACT, Australia
3 shared publications
Latest Publications
- 11 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
Heuristic grasping of convex objects using 3D imaging and tactile sensing in uncalibrated grasping scenarios
- 11 July 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 208
Heuristic grasping of convex objects using 3D imaging and tactile sensing in uncalibrated grasping scenarios
- 11 November 2021
- Published by Springer Nature in Biological Cybernetics
- Vol. 116 (1) , 69-79
On motion camouflage as proportional navigation
Motion camouflage is a stealth behaviour by which an insect can appear stationary at a fixed point while approaching or escaping another moving insect. Although several approaches have been proposed to generate motion camouflage in simulated and real agents, the exact mechanisms insects use to perform this complex behaviour are not well understood, especially considering their limited perceptual and computational resources. This paper sheds light on the possible underlying control mechanisms insect might use to generate motion camouflage, by training and analysing a series of motion camouflage controllers using reinforcement learning. We first investigate through simulations the most relevant information available to the insect that can be used to perform motion camouflage and analyse the learnt controllers. The results of this analysis drove us to hypothesise two simpler control mechanisms which, we show, can also generate motion camouflage. The proposed controllers are an extension of proportional navigation, another interception technique found in nature, and therefore, both animal behaviours seem to be connected. Motion camouflage can lead, among others, to novel approaches to closely observe animals in the wild, record sports events or gather information in military operations without being noticed.
- 11 November 2021
- Published by Springer Nature in Biological Cybernetics
- Vol. 116 (1) , 69-79
On motion camouflage as proportional navigation
Motion camouflage is a stealth behaviour by which an insect can appear stationary at a fixed point while approaching or escaping another moving insect. Although several approaches have been proposed to generate motion camouflage in simulated and real agents, the exact mechanisms insects use to perform this complex behaviour are not well understood, especially considering their limited perceptual and computational resources. This paper sheds light on the possible underlying control mechanisms insect might use to generate motion camouflage, by training and analysing a series of motion camouflage controllers using reinforcement learning. We first investigate through simulations the most relevant information available to the insect that can be used to perform motion camouflage and analyse the learnt controllers. The results of this analysis drove us to hypothesise two simpler control mechanisms which, we show, can also generate motion camouflage. The proposed controllers are an extension of proportional navigation, another interception technique found in nature, and therefore, both animal behaviours seem to be connected. Motion camouflage can lead, among others, to novel approaches to closely observe animals in the wild, record sports events or gather information in military operations without being noticed.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 705-713
Using Reinforcement Learning to Attenuate for Stochasticity in Robot Navigation Controllers
Braitenberg vehicles are bio-inspired controllers for sensor-based local navigation of wheeled robots that have been used in multiple real world robotic implementations. The common approach to implement such non-linear control mechanisms is through neural networks connecting sensing to motor action, yet tuning the weights to obtain appropriate closed-loop navigation behaviours can be very challenging. Standard approaches used hand tuned spiking or recurrent neural networks, or learnt the weights of feedforward networks using evolutionary approaches. Recently, Reinforcement Learning has been used to learn neural controllers for simulated Braitenberg vehicle 3a – a bio-inspired model of target seeking for wheeled robots – under the assumption of noiseless sensors. Real sensors, however, are subject to different levels of noise, and multiple works have shown that Braitenberg vehicles work even on outdoor robots, demonstrating that these control mechanisms work in harsh and dynamic environments. This paper shows that a robust neural controller for Braitenberg vehicle 3a can be learnt using policy gradient reinforcement learning in scenarios where sensor noise plays a non negligible role. The learnt controller is robust and tries to attenuate the effects of noise in the closed- loop navigation behaviour of the simulated stochastic vehicle. We compare the neural controller learnt using Reinforcement Learning with a simple hand tuned controller and show how the neural control mechanism outperforms a naïve controller. Results are illustrated through computer simulations of the closed-loop stochastic system.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 705-713
Using Reinforcement Learning to Attenuate for Stochasticity in Robot Navigation Controllers
Braitenberg vehicles are bio-inspired controllers for sensor-based local navigation of wheeled robots that have been used in multiple real world robotic implementations. The common approach to implement such non-linear control mechanisms is through neural networks connecting sensing to motor action, yet tuning the weights to obtain appropriate closed-loop navigation behaviours can be very challenging. Standard approaches used hand tuned spiking or recurrent neural networks, or learnt the weights of feedforward networks using evolutionary approaches. Recently, Reinforcement Learning has been used to learn neural controllers for simulated Braitenberg vehicle 3a – a bio-inspired model of target seeking for wheeled robots – under the assumption of noiseless sensors. Real sensors, however, are subject to different levels of noise, and multiple works have shown that Braitenberg vehicles work even on outdoor robots, demonstrating that these control mechanisms work in harsh and dynamic environments. This paper shows that a robust neural controller for Braitenberg vehicle 3a can be learnt using policy gradient reinforcement learning in scenarios where sensor noise plays a non negligible role. The learnt controller is robust and tries to attenuate the effects of noise in the closed- loop navigation behaviour of the simulated stochastic vehicle. We compare the neural controller learnt using Reinforcement Learning with a simple hand tuned controller and show how the neural control mechanism outperforms a naïve controller. Results are illustrated through computer simulations of the closed-loop stochastic system.
- 17 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive and Developmental Systems
- Vol. 13 (2) , 274-285
Open-Ended Continuous Learning of Compound Goals
- 17 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive and Developmental Systems
- Vol. 13 (2) , 274-285
Open-Ended Continuous Learning of Compound Goals
- Open Access
- 23 May 2019
- Published by Springer Nature in Nature Communications
- Vol. 10 (1) , 1-12
A neural circuit model of decision uncertainty and change-of-mind
Decision-making is often accompanied by a degree of confidence on whether a choice is correct. Decision uncertainty, or lack in confidence, may lead to change-of-mind. Studies have identified the behavioural characteristics associated with decision confidence or change-of-mind, and their neural correlates. Although several theoretical accounts have been proposed, there is no neural model that can compute decision uncertainty and explain its effects on change-of-mind. We propose a neuronal circuit model that computes decision uncertainty while accounting for a variety of behavioural and neural data of decision confidence and change-of-mind, including testable model predictions. Our theoretical analysis suggests that change-of-mind occurs due to the presence of a transient uncertainty-induced choice-neutral stable steady state and noisy fluctuation within the neuronal network. Our distributed network model indicates that the neural basis of change-of-mind is more distinctively identified in motor-based neurons. Overall, our model provides a framework that unifies decision confidence and change-of-mind.
- 23 May 2019
- Published by Springer Nature in Nature Communications
- Vol. 10 (1) , 1-12
Open AccessA neural circuit model of decision uncertainty and change-of-mind
Decision-making is often accompanied by a degree of confidence on whether a choice is correct. Decision uncertainty, or lack in confidence, may lead to change-of-mind. Studies have identified the behavioural characteristics associated with decision confidence or change-of-mind, and their neural correlates. Although several theoretical accounts have been proposed, there is no neural model that can compute decision uncertainty and explain its effects on change-of-mind. We propose a neuronal circuit model that computes decision uncertainty while accounting for a variety of behavioural and neural data of decision confidence and change-of-mind, including testable model predictions. Our theoretical analysis suggests that change-of-mind occurs due to the presence of a transient uncertainty-induced choice-neutral stable steady state and noisy fluctuation within the neuronal network. Our distributed network model indicates that the neural basis of change-of-mind is more distinctively identified in motor-based neurons. Overall, our model provides a framework that unifies decision confidence and change-of-mind.
- 2 January 2019
- Published by Springer Nature in Autonomous Robots
- Vol. 43 (7) , 1623-1637
Reliable robotic handovers through tactile sensing
- 2 January 2019
- Published by Springer Nature in Autonomous Robots
- Vol. 43 (7) , 1623-1637
Reliable robotic handovers through tactile sensing
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1394-1399
Bridging the Gap between Bio-Inspired Steering and Locomotion: A Braitenberg 3a Snake Robot
Braitenberg vehicles are simple models of animal motion towards, or away from, a stimulus (light, sound, chemicals, etc). They have been widely used in robotics to implement target reaching and avoidance behaviors based on different types of sensors. While the seminal work of Braitenberg used wheeled vehicles to illustrate the principles of animal steering, few attempts have been made at combining these steering level controllers with locomotion mechanism other than actuated wheels. This paper presents the first implementation of this biologically inspired steering controller in a snake-like robot with non-actuated wheels and actuated joints. The sinusoidal gait of the snake is modulated following the principles of the Braitenberg vehicle 3α using two sensors symmetrically located on the head. The effectiveness of this bio-inspired controller is shown through simulations where the snake orients its head and body with the direction of the stimulus gradient, and reaches the stimulus maximum within some range. This paper represents one of the first steps towards connecting bio-inspired sensor-based steering mechanisms and bio-inspired locomotion, and shows that existing theoretical results of Braitenberg vehicles with actuated wheels also apply to a snake-like robot with non-actuated wheels.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1394-1399
Bridging the Gap between Bio-Inspired Steering and Locomotion: A Braitenberg 3a Snake Robot
Braitenberg vehicles are simple models of animal motion towards, or away from, a stimulus (light, sound, chemicals, etc). They have been widely used in robotics to implement target reaching and avoidance behaviors based on different types of sensors. While the seminal work of Braitenberg used wheeled vehicles to illustrate the principles of animal steering, few attempts have been made at combining these steering level controllers with locomotion mechanism other than actuated wheels. This paper presents the first implementation of this biologically inspired steering controller in a snake-like robot with non-actuated wheels and actuated joints. The sinusoidal gait of the snake is modulated following the principles of the Braitenberg vehicle 3α using two sensors symmetrically located on the head. The effectiveness of this bio-inspired controller is shown through simulations where the snake orients its head and body with the direction of the stimulus gradient, and reaches the stimulus maximum within some range. This paper represents one of the first steps towards connecting bio-inspired sensor-based steering mechanisms and bio-inspired locomotion, and shows that existing theoretical results of Braitenberg vehicles with actuated wheels also apply to a snake-like robot with non-actuated wheels.
- Open Access
- 9 October 2018
- Published by Frontiers Media SA in Frontiers in Neurorobotics
- Vol. 12, 63
Intrinsic Rewards for Maintenance, Approach, Avoidance, and Achievement Goal Types
In reinforcement learning, reward is used to guide the learning process. The reward is often designed to be task-dependent, and it may require significant domain knowledge to design a good reward function. This paper proposes general reward functions for maintenance, approach, avoidance, and achievement goal types. These reward functions exploit the inherent property of each type of goal and are thus task-independent. We also propose metrics to measure an agent's performance for learning each type of goal. We evaluate the intrinsic reward functions in a framework that can autonomously generate goals and learn solutions to those goals using a standard reinforcement learning algorithm. We show empirically how the proposed reward functions lead to learning in a mobile robot application. Finally, using the proposed reward functions as building blocks, we demonstrate how compound reward functions, reward functions to generate sequences of tasks, can be created that allow the mobile robot to learn more complex behaviors.
- 9 October 2018
- Published by Frontiers Media SA in Frontiers in Neurorobotics
- Vol. 12, 63
Open AccessIntrinsic Rewards for Maintenance, Approach, Avoidance, and Achievement Goal Types
In reinforcement learning, reward is used to guide the learning process. The reward is often designed to be task-dependent, and it may require significant domain knowledge to design a good reward function. This paper proposes general reward functions for maintenance, approach, avoidance, and achievement goal types. These reward functions exploit the inherent property of each type of goal and are thus task-independent. We also propose metrics to measure an agent's performance for learning each type of goal. We evaluate the intrinsic reward functions in a framework that can autonomously generate goals and learn solutions to those goals using a standard reinforcement learning algorithm. We show empirically how the proposed reward functions lead to learning in a mobile robot application. Finally, using the proposed reward functions as building blocks, we demonstrate how compound reward functions, reward functions to generate sequences of tasks, can be created that allow the mobile robot to learn more complex behaviors.
- 1 August 2018
- Published by Elsevier in Robotics and Autonomous Systems
- Vol. 106, 130-139
Multimodal Material identification through recursive tactile sensing
- 1 August 2018
- Published by Elsevier in Robotics and Autonomous Systems
- Vol. 106, 130-139
Multimodal Material identification through recursive tactile sensing
- Open Access
Neural Circuit Mechanism of Decision Uncertainty and Change-of-Mind
Decision-making is often accompanied by a degree of confidence on whether a choice is correct. Decision uncertainty, or lack in confidence, may lead to change-of-mind. Studies have identified the behavioural characteristics associated with decision confidence or change-of-mind, and their neural correlates. Although several theoretical accounts have been proposed, there is no neural model that can compute decision uncertainty and explain its effects on change-of-mind. We propose a neuronal circuit model that computes decision uncertainty while accounting for a variety of behavioural and neural data of decision confidence and change-of-mind, including testable model predictions. Our theoretical analysis suggests that change-of-mind occurs due to the presence of a transient uncertainty-induced choice-neutral stable steady state and noisy fluctuation within the neuronal network. Our distributed network model indicates that the neural basis of change-of-mind is more distinctively identified in motor-based neurons. Overall, our model provides a framework that unifies decision confidence and change-of-mind.
Open AccessNeural Circuit Mechanism of Decision Uncertainty and Change-of-Mind
Decision-making is often accompanied by a degree of confidence on whether a choice is correct. Decision uncertainty, or lack in confidence, may lead to change-of-mind. Studies have identified the behavioural characteristics associated with decision confidence or change-of-mind, and their neural correlates. Although several theoretical accounts have been proposed, there is no neural model that can compute decision uncertainty and explain its effects on change-of-mind. We propose a neuronal circuit model that computes decision uncertainty while accounting for a variety of behavioural and neural data of decision confidence and change-of-mind, including testable model predictions. Our theoretical analysis suggests that change-of-mind occurs due to the presence of a transient uncertainty-induced choice-neutral stable steady state and noisy fluctuation within the neuronal network. Our distributed network model indicates that the neural basis of change-of-mind is more distinctively identified in motor-based neurons. Overall, our model provides a framework that unifies decision confidence and change-of-mind.