Skip to main content

Tony Pipe

University of the West of England, Bristol Robotics Lab, Faculty Member

Followers

60

Following

18

Co-authors

16

Public Views

Address: Bristol, Bristol, City of, United Kingdom

less

Interests

Uploads

Papers by Tony Pipe

Towards Safe Human-Robot Interaction

by Elena Corina Grigore, Kerstin Eder, and Tony Pipe

Lecture Notes in Computer Science, 2011

The development of human-assistive robots challenges engineering and introduces new ethical and l... more The development of human-assistive robots challenges engineering and introduces new ethical and legal issues. One fundamental concern is whether human-assistive robots can be trusted. Essential components of trustworthiness are usefulness and safety; both have to be demonstrated before such robots could stand a chance of passing product certification. This paper describes the setup of an environment to investigate safety and liveness aspects in the context of human-robot interaction. We present first insights into setting up and testing a humanrobot interaction system in which the role of the robot is that of serving drinks to a human. More specifically, we use this system to investigate when the right time is for the robot to release the drink such that the action is both safe and useful. We briefly outline follow-on research that uses the safety and liveness properties of this scenario as specification.

Self-adaptive context aware audio localization for robots using parallel cerebellar models

by Emanuele Lindo Secco and Tony Pipe

, Towards Autonomous Robotic Systems (TAROS), Lecture Notes in Computer Sciences, Chapter 6, 1-13, 2017, 2017

An audio sensor system is presented that uses multiple cerebellar models to determine the acousti... more An audio sensor system is presented that uses multiple cerebellar models to determine the acoustic environment in which a robot is operating, allowing the robot to select appropriate models to calibrate its audio-motor map for the detected environment. The use of the adaptive filter model of the cerebellum in a variety of robotics applications has demonstrated the utility of the so-called cerebellar chip. This paper combines the notion of cerebellar calibration of a distorted audio-motor map with the use of multiple parallel models to predict the context (acoustic environment) within which the robot is operating. The system was able to correctly predict seven different acoustic contexts in almost 70% of cases tested.

Ethical Issues for Robotics and Autonomous Systems

UKRAS White Papers, 2019

Longitudinal Driving Skills Transfer from Driver to Smart Vehicle

2021 26th International Conference on Automation and Computing (ICAC)

An Agency-Directed Approach to Test Generation for Simulation-based Autonomous Vehicle Verification

2020 IEEE International Conference On Artificial Intelligence Testing (AITest)

Novel bio-inspired self-repair algorithm for evolvable fault tolerant hardware systems

Proceedings of the 11th Annual Conference Companion, Jul 8, 2009

Abstract This paper investigates and presents a novel self-repair algorithm, based on a prokaryot... more

Biologically Inspired Solutions for Compliant Limb UGVs

Whilst the energy efficiency of wheeled vehicles on smooth ground is undeniable, negotiating brok... more Whilst the energy efficiency of wheeled vehicles on smooth ground is undeniable, negotiating broken or complex terrain still presents a significant problem. This paper discusses the advances made in producing a legged UGV platform; specifically, we focus on three areas which have been identified as being key to developing a viable solution. These elements are fundamentally coupled as the breakdown of one will result in the failure of the entire system. Thus our guiding design principle of biological inspiration must be grounded with a thorough understanding of the limitations imposed by current mechanical technologies. The autonomous nature of the system therefore identifies weight reduction as a key aspect in the system level decision making process.

Learning to predict with the Delayed Action Classifier System

Proceedings of the 2003 IEEE International Symposium on Intelligent Control ISIC-03, 2003

This paper describes an extended version of the Delayed Action Classifier System (DACS), a rule-b... more This paper describes an extended version of the Delayed Action Classifier System (DACS), a rule-based system which employs the genetic algorithm to discover temporal rules for learning in environments with temporal structure. The extended version of DACS (called DACS2) is described and experimentally evaluated. A mathematical analysis is provided which offers a theoretical justification of the advantages of DACS compared to non-temporal classifier systems. Areas for further development are briefly discussed and possible applications of the system in the field of intelligent control are suggested.

Artificial Innate Immune System: An Instant Defence Layer of Embryonics

Lecture Notes in Computer Science, 2004

... In 1998, Cesar Ortega proposed another embryonic model [22] also based on MUXTREE principles ... more

Investigating Predictive Model-Based Robot Behaviour Adaptation

OnlineResource4

Biomimetic Tactile Sensing Capsule

Lecture Notes in Computer Science, 2015

A Novel Q-Learning Based Adaptive Optimal Controller Implementation for a Humanoid Robotic Arm

Proceedings of the 18th IFAC World Congress, 2011

This paper presents the implementation of a novel model-free Q-learning based discrete adaptive o... more This paper presents the implementation of a novel model-free Q-learning based discrete adaptive optimal controller for a humanoid robotic arm. The controller uses a novel adaptive dynamic programming (ADP) reinforcement learning (RL) approach to develop an optimal policy on-line. This is in contrast with the other optimal control design techniques which are carried out off-line and need full information of the system dynamics. The RL tracking controller was implemented for two links (shoulder flexion and elbow flexion joints) of the arm of the humanoid Bristol-Elumotion-Robotic-Torso II (BERT II) torso. The constrained case (joint limits) of the RL scheme was tested for a single link (elbow flexion) of the BERT II arm by modifying the cost function to deal with the extra nonlinearity due to the joint constraint. 1 BERT II has been developed in a collaboration of the Bristol Robotics Laboratory and Elumotion Ltd.

Investigating real world issues in Human Robot Interaction: Physical and Cognitive solutions for a safe robotic system

Steps Forward to Evolve Bio-inspired Embryonic Cell-Based Electronic Systems

Lecture Notes in Computer Science, 2007

EHW is the acronym used to denote an emerging and relatively new research field in digital hardwa... more EHW is the acronym used to denote an emerging and relatively new research field in digital hardware design; it stands for Evolvable Hardware. This technique has attracted many researchers since the 1990’s. EHW aims at an automatic design and optimisation of a reconfigurable hardware system using Evolutionary Algorithms (EAs), such as Genetic Algorithms, Genetic programming etc. This article is published as part of a three years research project. The objective of this project is to employ the above method on a target specific hardware, the Embryonics Hardware System. The latter requires large hardware resources. Thus, in this project, EAs will be used to evolve the Embryonics Hardware System to discover novel design with reduced complexity. The new design must first ensure the correct functionality. Hence to verify the concept of Evolvable Hardware, the authors, in this paper, focus on the design of relatively simple combinatorial logic circuits using Genetic Algorithms with multi-objective fitness.

Whiskered robots

Scholarpedia, 2015

Towards a variable compliance hydrostatic skeleton inspired gripper

2012 4th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob), 2012

ABSTRACT Variable compliance is a subject of great interest in the robotics community, particular... more ABSTRACT Variable compliance is a subject of great interest in the robotics community, particularly in the area of Human-Robot interaction. In this work, variable compliance is achieved in a novel grasping device inspired by the hydrostatic skeletons of invertebrates. This is accomplished passively, relying on material properties and innovative prototype design. The design will be described and its behavior evaluated through a series of experiments. It will also be shown that a prototype with a very simple design is able to grasp the object with little active control.

The BERT2 infrastructure: An integrated system for the study of human-robot interaction

2010 10th IEEE-RAS International Conference on Humanoid Robots, 2010

AbstractBristol Elumotion Robot Torso Version 2 (BERT2) is a humanoid robot currently in develop... more

Adjustable Torque Limit Variable Compliance Decoupling Joint

One of the most challenging features of service robotics is physical Human-Robot Interaction (pHR... more One of the most challenging features of service robotics is physical Human-Robot Interaction (pHRI), [1] because of the safety hazards involved in interaction. In order to increase safety, mechanics, software and electronics issues should be addressed but in this paper our focus will be on the mechanical structure. Such choice is motivated by the fact that systems which rely on their physical structure rather than on active control, those which have a passive compliance system, can provide faster and more reliable responses for dynamic impacts than an active one[2]. Also, a passive system has claims to inherent safety, which ensures that a system is behaving in a safe way even if sensors and control fail. The inherent-safety approach dictates the use of flexible, shock-absorbing joints that can sustain sudden collisions. The designed joint, Figure 1, passively decouples the arm link from the motor in case of a collision. Its design is inspired by the work of Park et al. [2] but has a major advantage over it: its torque threshold can be actively regulated depending on the operational situation. For example, it would be possible to stiffen the joint when the arm has to cope with a high load in a given task but the decoupling joint could be made more compliant in the presence of fragile obstacles (objects or humans), reducing the harm potential. However, once the torque threshold has been adjusted, the decoupling joint resumes its completely passive behaviour. The variable torque threshold is obtained by changing angle theta, equation 1. This is obtained by moving the "arms" of an ad hoc designed and built system which transfers the movement given by the linear actuator through the decoupling joint.

Safe Adaptive Compliance Control of a Humanoid Robotic Arm with Anti-Windup Compensation and Posture Control

International Journal of Social Robotics, 2010

Safety is very important for physical human-robot interaction. Compliance control can solve an im... more Safety is very important for physical human-robot interaction. Compliance control can solve an important aspect of the safety problem by dealing with impact and other forces arising during close contact between humans and robots. An adaptive compliance model reference controller was implemented in real-time on a 4 degrees of freedom (DOF) humanoid robotic arm in Cartesian space. The robot manipulator has been controlled in such a way as to follow the compliant passive behaviour of a reference mass-springdamper system model subject to an externally sensed force. The redundant DOF were used to control the robot motion in a human-like pattern via minimization of effort, a function of gravity. Associated actuator saturation issues were addressed by incorporating a novel anti-windup (AW) compensator originally developed for a neural network scheme. The controller was simulated for a robotic arm representing the Bristol-Elumotion-Robotic-Torso II (BERT II) and then tested on the real BERT II arm. BERT II has been developed in collaboration by Bristol Robotics Laboratory and Elumotion Ltd.

Towards Safe Human-Robot Interaction

by Elena Corina Grigore, Kerstin Eder, and Tony Pipe

Lecture Notes in Computer Science, 2011

The development of human-assistive robots challenges engineering and introduces new ethical and l... more The development of human-assistive robots challenges engineering and introduces new ethical and legal issues. One fundamental concern is whether human-assistive robots can be trusted. Essential components of trustworthiness are usefulness and safety; both have to be demonstrated before such robots could stand a chance of passing product certification. This paper describes the setup of an environment to investigate safety and liveness aspects in the context of human-robot interaction. We present first insights into setting up and testing a humanrobot interaction system in which the role of the robot is that of serving drinks to a human. More specifically, we use this system to investigate when the right time is for the robot to release the drink such that the action is both safe and useful. We briefly outline follow-on research that uses the safety and liveness properties of this scenario as specification.

Self-adaptive context aware audio localization for robots using parallel cerebellar models

by Emanuele Lindo Secco and Tony Pipe

, Towards Autonomous Robotic Systems (TAROS), Lecture Notes in Computer Sciences, Chapter 6, 1-13, 2017, 2017

An audio sensor system is presented that uses multiple cerebellar models to determine the acousti... more An audio sensor system is presented that uses multiple cerebellar models to determine the acoustic environment in which a robot is operating, allowing the robot to select appropriate models to calibrate its audio-motor map for the detected environment. The use of the adaptive filter model of the cerebellum in a variety of robotics applications has demonstrated the utility of the so-called cerebellar chip. This paper combines the notion of cerebellar calibration of a distorted audio-motor map with the use of multiple parallel models to predict the context (acoustic environment) within which the robot is operating. The system was able to correctly predict seven different acoustic contexts in almost 70% of cases tested.

Ethical Issues for Robotics and Autonomous Systems

UKRAS White Papers, 2019

Longitudinal Driving Skills Transfer from Driver to Smart Vehicle

2021 26th International Conference on Automation and Computing (ICAC)

An Agency-Directed Approach to Test Generation for Simulation-based Autonomous Vehicle Verification

2020 IEEE International Conference On Artificial Intelligence Testing (AITest)

Novel bio-inspired self-repair algorithm for evolvable fault tolerant hardware systems

Proceedings of the 11th Annual Conference Companion, Jul 8, 2009

Abstract This paper investigates and presents a novel self-repair algorithm, based on a prokaryot... more

Biologically Inspired Solutions for Compliant Limb UGVs

Whilst the energy efficiency of wheeled vehicles on smooth ground is undeniable, negotiating brok... more Whilst the energy efficiency of wheeled vehicles on smooth ground is undeniable, negotiating broken or complex terrain still presents a significant problem. This paper discusses the advances made in producing a legged UGV platform; specifically, we focus on three areas which have been identified as being key to developing a viable solution. These elements are fundamentally coupled as the breakdown of one will result in the failure of the entire system. Thus our guiding design principle of biological inspiration must be grounded with a thorough understanding of the limitations imposed by current mechanical technologies. The autonomous nature of the system therefore identifies weight reduction as a key aspect in the system level decision making process.

Learning to predict with the Delayed Action Classifier System

Proceedings of the 2003 IEEE International Symposium on Intelligent Control ISIC-03, 2003

This paper describes an extended version of the Delayed Action Classifier System (DACS), a rule-b... more This paper describes an extended version of the Delayed Action Classifier System (DACS), a rule-based system which employs the genetic algorithm to discover temporal rules for learning in environments with temporal structure. The extended version of DACS (called DACS2) is described and experimentally evaluated. A mathematical analysis is provided which offers a theoretical justification of the advantages of DACS compared to non-temporal classifier systems. Areas for further development are briefly discussed and possible applications of the system in the field of intelligent control are suggested.

Artificial Innate Immune System: An Instant Defence Layer of Embryonics

Lecture Notes in Computer Science, 2004

... In 1998, Cesar Ortega proposed another embryonic model [22] also based on MUXTREE principles ... more

Investigating Predictive Model-Based Robot Behaviour Adaptation

OnlineResource4

Biomimetic Tactile Sensing Capsule

Lecture Notes in Computer Science, 2015

A Novel Q-Learning Based Adaptive Optimal Controller Implementation for a Humanoid Robotic Arm

Proceedings of the 18th IFAC World Congress, 2011

This paper presents the implementation of a novel model-free Q-learning based discrete adaptive o... more This paper presents the implementation of a novel model-free Q-learning based discrete adaptive optimal controller for a humanoid robotic arm. The controller uses a novel adaptive dynamic programming (ADP) reinforcement learning (RL) approach to develop an optimal policy on-line. This is in contrast with the other optimal control design techniques which are carried out off-line and need full information of the system dynamics. The RL tracking controller was implemented for two links (shoulder flexion and elbow flexion joints) of the arm of the humanoid Bristol-Elumotion-Robotic-Torso II (BERT II) torso. The constrained case (joint limits) of the RL scheme was tested for a single link (elbow flexion) of the BERT II arm by modifying the cost function to deal with the extra nonlinearity due to the joint constraint. 1 BERT II has been developed in a collaboration of the Bristol Robotics Laboratory and Elumotion Ltd.

Investigating real world issues in Human Robot Interaction: Physical and Cognitive solutions for a safe robotic system

Steps Forward to Evolve Bio-inspired Embryonic Cell-Based Electronic Systems

Lecture Notes in Computer Science, 2007

EHW is the acronym used to denote an emerging and relatively new research field in digital hardwa... more EHW is the acronym used to denote an emerging and relatively new research field in digital hardware design; it stands for Evolvable Hardware. This technique has attracted many researchers since the 1990’s. EHW aims at an automatic design and optimisation of a reconfigurable hardware system using Evolutionary Algorithms (EAs), such as Genetic Algorithms, Genetic programming etc. This article is published as part of a three years research project. The objective of this project is to employ the above method on a target specific hardware, the Embryonics Hardware System. The latter requires large hardware resources. Thus, in this project, EAs will be used to evolve the Embryonics Hardware System to discover novel design with reduced complexity. The new design must first ensure the correct functionality. Hence to verify the concept of Evolvable Hardware, the authors, in this paper, focus on the design of relatively simple combinatorial logic circuits using Genetic Algorithms with multi-objective fitness.

Whiskered robots

Scholarpedia, 2015

Towards a variable compliance hydrostatic skeleton inspired gripper

2012 4th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob), 2012

ABSTRACT Variable compliance is a subject of great interest in the robotics community, particular... more ABSTRACT Variable compliance is a subject of great interest in the robotics community, particularly in the area of Human-Robot interaction. In this work, variable compliance is achieved in a novel grasping device inspired by the hydrostatic skeletons of invertebrates. This is accomplished passively, relying on material properties and innovative prototype design. The design will be described and its behavior evaluated through a series of experiments. It will also be shown that a prototype with a very simple design is able to grasp the object with little active control.

The BERT2 infrastructure: An integrated system for the study of human-robot interaction

2010 10th IEEE-RAS International Conference on Humanoid Robots, 2010

AbstractBristol Elumotion Robot Torso Version 2 (BERT2) is a humanoid robot currently in develop... more

Adjustable Torque Limit Variable Compliance Decoupling Joint

One of the most challenging features of service robotics is physical Human-Robot Interaction (pHR... more One of the most challenging features of service robotics is physical Human-Robot Interaction (pHRI), [1] because of the safety hazards involved in interaction. In order to increase safety, mechanics, software and electronics issues should be addressed but in this paper our focus will be on the mechanical structure. Such choice is motivated by the fact that systems which rely on their physical structure rather than on active control, those which have a passive compliance system, can provide faster and more reliable responses for dynamic impacts than an active one[2]. Also, a passive system has claims to inherent safety, which ensures that a system is behaving in a safe way even if sensors and control fail. The inherent-safety approach dictates the use of flexible, shock-absorbing joints that can sustain sudden collisions. The designed joint, Figure 1, passively decouples the arm link from the motor in case of a collision. Its design is inspired by the work of Park et al. [2] but has a major advantage over it: its torque threshold can be actively regulated depending on the operational situation. For example, it would be possible to stiffen the joint when the arm has to cope with a high load in a given task but the decoupling joint could be made more compliant in the presence of fragile obstacles (objects or humans), reducing the harm potential. However, once the torque threshold has been adjusted, the decoupling joint resumes its completely passive behaviour. The variable torque threshold is obtained by changing angle theta, equation 1. This is obtained by moving the "arms" of an ad hoc designed and built system which transfers the movement given by the linear actuator through the decoupling joint.

Safe Adaptive Compliance Control of a Humanoid Robotic Arm with Anti-Windup Compensation and Posture Control

International Journal of Social Robotics, 2010

Safety is very important for physical human-robot interaction. Compliance control can solve an im... more Safety is very important for physical human-robot interaction. Compliance control can solve an important aspect of the safety problem by dealing with impact and other forces arising during close contact between humans and robots. An adaptive compliance model reference controller was implemented in real-time on a 4 degrees of freedom (DOF) humanoid robotic arm in Cartesian space. The robot manipulator has been controlled in such a way as to follow the compliant passive behaviour of a reference mass-springdamper system model subject to an externally sensed force. The redundant DOF were used to control the robot motion in a human-like pattern via minimization of effort, a function of gravity. Associated actuator saturation issues were addressed by incorporating a novel anti-windup (AW) compensator originally developed for a neural network scheme. The controller was simulated for a robotic arm representing the Bristol-Elumotion-Robotic-Torso II (BERT II) and then tested on the real BERT II arm. BERT II has been developed in collaboration by Bristol Robotics Laboratory and Elumotion Ltd.

Whiskerbot: A Robotic Active Touch System Modeled on the Rat Whisker Sensory System

by Tony J Prescott and Tony Pipe

Adaptive Behavior, 2007

... that interfaces the hardware instantiated neural models in the FPGA with the software impleme... more

Towards biomimetic vibrissal tactile sensing for robot exploration, navigation, and object recognition in hazardous environments

by S. Anderson, Tony J Prescott, and Tony Pipe

Contact type dependency of texture classification in a whiskered mobile robot

by Tony J Prescott, Tony Pipe, and Benjamin Mitchinson

Autonomous Robots, 2009

Actuated artificial whiskers modeled on rat macrovibrissae can provide effective tactile sensor s... more Actuated artificial whiskers modeled on rat macrovibrissae can provide effective tactile sensor systems for autonomous robots. This article focuses on texture classification using artificial whiskers and addresses a limitation of previous studies, namely, their use of whisker deflection signals obtained under relatively constrained experimental conditions. Here we consider the classification of signals obtained from a whiskered robot required to explore different surface textures from a range of orientations and distances. This procedure resulted in a variety of deflection signals for any given texture. Using a standard Gaussian classifier we show, using both hand-picked features and ones derived from studies of rat vibrissal processing, that a robust rough-smooth discrimination is achievable without any knowledge of how the whisker interacts with the investigated object. On the other hand, finer discriminations appear to require knowledge of the target's relative position and/or of the manner in which the whisker contact its surface. (146 words)

The robot vibrissal system: Understanding mammalian sensorimotor co-ordination through biomimetics

by Tony J Prescott, Stuart Wilson, Tony Pipe, and Charles Sullivan

We consider the problem of sensorimotor coordination in mammals through the lens of vibrissal tou... more We consider the problem of sensorimotor coordination in mammals through the lens of vibrissal touch, and via the methodology of embodied computational neuroscience—using biomimetic robots to synthesize and investigate models of mammalian brain architecture. The chapter focuses on five major brain subsystems and their likely role in vibrissal system function—superior colliculus, basal ganglia, somatosensory cortex, cerebellum, and hippocampus. With respect to each of these we demonstrate how embodied modelling has helped elucidate their likely function in the brain of awake behaving animals. We also demonstrate how the appropriate coordination of these subsystems , with a model of brain architecture, can give rise to integrated behaviour in a lifelike whiskered robot.