Estela Bicho

Target representation on an autonomous vehicle with low-level sensors

How can low-level autonomous robots with only very simple sensor systems be endowed with cognitive capabilities? Specifically, we consider a system that uses seven infrared sensors and five microphones to avoid obstacles and acquire sound targets. The cognitive abilities of the vehicle consist of representing the direction in which a sound source lies. This representation supports target detection, estimation of target direction, selection of one out of multiple-detected targets, storage of target direction in short-term memory, continuous updating of memory, and deletion of memorized target information after a characteristic delay. We show that the dynamic approach (attractor dynamics) employed to control the motion of the robot can be extended to the level of representation by using dynamic neural fields to interpolate sensory information. We show how the system stabilizes decisions in the presence of multivalue sensorial information and activates and deactivates memory. Smooth integration of this target representation with target acquisition, in the form of phonotaxis, and obstacle avoidance is demonstrated.

The dynamic neural field approach to cognitive robotics

This tutorial presents an architecture for autonomous robots to generate behavior in joint action tasks. To efficiently interact with another agent in solving a mutual task, a robot should be endowed with cognitive skills such as memory, decision making, action understanding and prediction. The proposed architecture is strongly inspired by our current understanding of the processing principles and the neuronal circuitry underlying these functionalities in the primate brain. As a mathematical framework, we use a coupled system of dynamic neural fields, each representing the basic functionality of neuronal populations in different brain areas. It implements goal-directed behavior in joint action as a continuous process that builds on the interpretation of observed movements in terms of the partners action goal. We validate the architecture in two experimental paradigms: (1) a joint search task; (2) a reproduction of an observed or inferred end state of a grasping-placing sequence. We also review some of the mathematical results about dynamic neural fields that are important for the implementation work.

The dynamic approach to autonomous robotics demonstrated on a low-level vehicle platform

The dynamic approach proposes a set of concepts with the help of which autonomous systems can be specified and designed. While the approach builds systems from elementary behaviors driven by behavior-specific sensory information, it also represents behaviors internally in terms of the state of dynamical systems, thus positioning itself somewhere between classical and behavior-based approaches. This paper demonstrates that the dynamic approach lends itself naturally to implementation on computationally weak platforms working with very low-level sensory information. Obstacle avoidance and target acquisition are implemented on a micro-controller based vehicle equipped with only five infra-red detectors and two photoresistors. We show how theoretical design, software simulation, and hardware implementation are enchained effortlessly. The resulting behavior is particularly smooth and requires no parameter optimization. As a technical novelty we demonstrate the integration of dynamics at two different levels of temporal derivative.

Formation control for multiple mobile robots: a non-linear attractor dynamics approach

In this paper we focus on modelling formations of non-holonomic mobile robots using non-linear attractor dynamics (see video). The benefit is that the behavior of each robot is generated by time series of asymptotically stable states, which therefore contribute to the robustness against environmental perturbations. This study extends our previous work [S Monteiro et al., 2002]. Here we develop a set of decentralized and distributed basic control architectures that allows each robot to maintain a desired pose within a formation and to enable changes in the shape of the formation which are necessary to avoid obstacles. Simulation results, for teams of four and six mobile robots driving in cluttered and unknown environments, while simultaneously trying to drive in line, column, square, diamond and hexagon are presented. We explain how this approach naturally extends to larger teams of robots.

Integrating verbal and nonverbal communication in a dynamic neural field architecture for human–robot interaction

How do humans coordinate their intentions, goals and motor behaviors when performing joint action tasks? Recent experimental evidence suggests that resonance processes in the observers motor system are crucially involved in our ability to understand actions of others’, to infer their goals and even to comprehend their action-related language. In this paper, we present a control architecture for human–robot collaboration that exploits this close perception-action linkage as a means to achieve more natural and efficient communication grounded in sensorimotor experiences. The architecture is formalized by a coupled system of dynamic neural fields representing a distributed network of neural populations that encode in their activation patterns goals, actions and shared task knowledge. We validate the verbal and nonverbal communication skills of the robot in a joint assembly task in which the human–robot team has to construct toy objects from their components. The experiments focus on the robots capacity to anticipate the users needs and to detect and communicate unexpected events that may occur during joint task execution.

Attractor dynamics approach to formation control: theory and application

In this paper we show how non-linear attractor dynamics can be used as a framework to control teams of autonomous mobile robots that should navigate according to a predefined geometric formation. The environment does not need to be known a priori and may change over time. Implicit to the control architecture are some important features such as establishing and moving the formation, split and join of formations (when necessary to avoid obstacles). Formations are defined by a formation matrix. By manipulating this formation matrix it is also possible to switch formations at run time. Examples of simulation results and implementations with real robots (teams of Khepera robots and medium size mobile robots), demonstrate formation switch, static and dynamic obstacle avoidance and split and join formations without the need for any explicit coordination scheme. Robustness against environmental perturbations is intrinsically achieved because the behaviour of each robot is generated as a time series of asymptotically stable states, which contribute to the asymptotic stability of the overall control system.

Attractor dynamics generates robot formation: from theory to implementation

We show how non-linear attractor dynamics can be used to implement robot formations in unknown environments. The desired formation geometry is given through a matrix where the parameters in each line (its leader, desired distance and relative orientation to the leader) define the desired pose of a robot in the formation. The parameter values are then used to shape the vector fields of the dynamical systems that generate values for the control variables (i.e. heading direction and path velocity). Then these dynamical systems are tuned such that the control variables are always very close to one of the resultant attractors. The advantage is that the systems are more robust against perturbations because the behavior is generated as a time series of asymptotically stable states. Experimental results (with three Khepera robots) demonstrate the ability of the team to create and stabilize the formation, as well as avoiding obstacles. Flexibility is achieved in that as the senses world changes, the systems may change their planning solutions continuously but also discontinuously (tuning the formation versus split to avoid obstacle).

Robot formations: Robots allocation and leader-follower pairs

In this paper we focus on the problem of assigning robots to places in a desired formation, considering random initial locations of the robots. Since we use a leader-follower strategy, we also address the task of choosing the leader to each follower. The result is a formation matrix that describes the relation between the robots and the desired formation shape. Simple algorithms are defined, that are based on the minimization of the distances of robots to places in the formation. All these algorithms are implemented in a decentralized way. We assume that communication is possible, but the requirements are of very-low bandwidth.

Object transportation by multiple mobile robots controlled by attractor dynamics: theory and implementation

Dynamical systems theory is used as a theoretical language and tool to design a distributed control architecture for teams of mobile robots, that must transport a large object and simultaneously avoid collisions with (either static or dynamic) obstacles. Here we demonstrate in simulations and implementations in real robots that it is possible to simplify the architectures presented in previous work and to extend the approach to teams of n robots. The robots have no prior knowledge of the environment. The motion of each robot is controlled by a time series of asymptotical stable states. The attractor dynamics permits the integration of information from various sources in a graded manner. As a result, the robots show a strikingly smooth an stable team behaviour.

Target position estimation, target acquisition, and obstacle avoidance

How can low-level autonomous robots with only very simple sensor systems be endowed with cognitive capabilities? Specifically, we consider a system which uses 5 infra-red sensors and 3 light-dependent resistors to acquire targets and avoid obstacles. How can the system be endowed with a continuous representation of target information, complete with subsymbolic memory and a memory decay process? We show that the dynamic approach employed to control the motion of the robot can be extended to the level of representation, if dynamic (neural) fields are used to interpolate sensory information. We show how the system stabilizes the decision, and activates and deactivates memory. Smooth integration of this dynamic target representation with target acquisition and obstacle avoidance is demonstrated.