Behavior analysis and training-a methodology for behavior engineering

Colombetti, Marco; Dorigo, Marco; Borghi, G.

doi:10.1109/3477.499789

Cited by 79 publications

(35 citation statements)

References 15 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Here we shall brie y sketch the ordering of decisions according to the BAT methodology (Colombetti, Dorigo, & Borghi, 1996Dorigo & Colombetti, 1998. The rst step in the development process is a su ciently detailed description of the robot, the environment, and the target behavior, that is, the task that the robot is intended to carry out.…”

Section: The Bat Methodologymentioning

confidence: 99%

“…In this section we apply the BAT methodology (Colombetti et al, 1996Dorigo & Colombetti, 1998 described in the previous section 2. In particular, we r u n a n experiment aimed at demonstrating two aspects of incremental robot shaping: incremental evolution to accommodate hardware and task modi cations, and automatic local adaptation of an individual module to changed environmental conditions.…”

Section: Evolution and Shapingmentioning

confidence: 99%

See 1 more Smart Citation

Incremental Robot Shaping

Urzelai¹,

Floreano²,

Dorigo³

et al. 1998

Connection Science

View full text Add to dashboard Cite

We propose a modular architecture for autonomous robots which allows for the implementation of basic behavioral modules by both programming and training, and accommodates for an evolutionary development o f t h e i n terconnections among modules. This architecture can implement highly complex controllers and allows for incremental shaping of the robot behavior. Our proposal is exempli ed and evaluated experimentally through a number of mobile robotic tasks involving exploration, battery recharging and object manipulation. Modern approaches to behavior engineering of autonomous robots have stressed the importance of modular and distributed architectures composed of simple and interconnected elements (Brooks, 1990Dorigo & Schnepf, 1991, 1993 Dorigo & C o l o m betti, 1994 where each component has full or partial access to sensory data and can a ect the actions taken by the robot. Distributed modular control has several potential advantages: it is an open system, it is intrinsically robust to local failures, and it is suitable for gradual \shaping", that is, incremental training of independent behavioral competencies (Dorigo & Colombetti, 1998). With modular architectures, relatively complex behavioral patterns can be built bottom-up from a set of simple basic behaviors. Two aspects are of key importance for the success of such a n a p p r o a c h: (i) the set of basic behaviors, A preliminary version of this paper was published in the proceedings of the Genetic Programming Conference, Madison, Wisconsin, 1998. 1 Incremental Robot Shaping 2 and (ii) the mutual interactions among them. As regards the rst point, the choice of which behavioral modules to assume as basic is typically made by a human designer. Once such a choice has been made, however, it is often feasible to use machine learning techniques as an aid to the implementation of the basic modules. Machine learning methods can also be exploited to develop the interactions among modules. Incremental Robot ShapingA problem which is often di cult to solve is nding the optimal balance between human design and the use of machine learning techniques. To n d a reasonable solution, one should always have a clear idea of why learning is used in a speci c application (Floreano, 1997). In fact, a robot's ability to learn its own behavior can be exploited to: (i) cut development costs by relieving human designers of part of their burden (ii) bypass the practical impossibility to completely describe the robot's environment and task a priori (iii) endow the robot with capacities for self-adaptation, which m a y p l a y an essential role in both optimizing behavior with respect to some performance measure, and in coping with unforeseen changes in the environment.However, to exploit machine learning techniques at their best, the whole development activity has to be conceived and organized in the appropriate way. In the course of our research, we have developed a methodology to assist an engineer in the process of designing and training an autonomous robot. The...

show abstract

Section: The Bat Methodologymentioning

confidence: 99%

Section: Evolution and Shapingmentioning

confidence: 99%

Incremental Robot Shaping

Urzelai¹,

Floreano²,

Dorigo³

et al. 1998

Connection Science

View full text Add to dashboard Cite

show abstract

“…The concept of learning based on behaviours instead of low level actuator control was proposed by Matarić in its PhD thesis [15], and later on this approach was adopted by other authors [16,17]. Dorigo and Colombetti combined RL and Classifier Systems to produce highly autonomous robots [18], based on a formalisation of the behavioural approach into a methodology similar to software engineering [19]. The idea of using behaviours not only for state space reduction but also as practical requirements for the safeness of the learning processes was proposed by Millan [20], who devised a connectionist architecture capable of combining fast learning from basic reflexes and high tolerance to sensor noise.…”

Section: Related Literaturementioning

confidence: 99%

Experiments in Robot Control for an Instance-Based Reinforcement Learning Algorithm based on Prior Information

Ribeiro¹,

Hemerly²

2016

Anais Do 4. Congresso Brasileiro De Redes Neurais

View full text Add to dashboard Cite

show abstract

“…As there is no principled methodology for creating the behaviour architecture, the designer has to use a laborious process of experimentation, trial and error to engineer the agent, in order to achieve an "eligible" system performance ,Rosenblatt and Payton (6). As an example, Behaviour Analysis and Training (BAT) methodology, Colombetti et al (14) assumes that behaviours are allocated to behaviour al modules by design, but the function of each module is developed by machine learning techniques. Although, it is possible to speed up training through the use of simulators, still the training phase can be an extremely expensive task.…”

Section: The Subsumption Architecturementioning

confidence: 99%

Exploiting ambient information into reactive agent architectures

Stamatis

Zaharakis

Kameas

2006

2nd IET International Conference on Intelligent Environments (IE 06)

View full text Add to dashboard Cite

This work presents an implementation of the subsumption architecture to agent individuals summing up a multi-agent system approach for the scale formation problem in secondary oil production. This approach is evaluated through a 3D simulator and some known issues concerning the architecture limitations are revealed and discussed. Then, the paper proposes a new hybrid architecture in order to minimize or even overcome these drawbacks.

show abstract

Behavior analysis and training-a methodology for behavior engineering

Cited by 79 publications

References 15 publications

Incremental Robot Shaping

Incremental Robot Shaping

Experiments in Robot Control for an Instance-Based Reinforcement Learning Algorithm based on Prior Information

Exploiting ambient information into reactive agent architectures

Contact Info

Product

Resources

About