Hardware-Based Particle Filter with Evolutionary Resampling Stage

Autonomous systems require, in most of the cases, reasoning and decision-making capabilities. Moreover, the decision process has to occur in real time. Real-time computing means that every situation or event has to have an answer before a temporal deadline. In complex applications, these deadlines are usually in the order of milliseconds or even microseconds if the application is very demanding. In order to comply with these timing requirements, computing tasks have to be performed as fast as possible. The problem arises when computations are no longer simple, but very time-consuming operations. A good example can be found in autonomous navigation systems with visual-tracking submodules where Kalman filtering is the most extended solution. However, in recent years, some interesting new approaches have been developed. Particle filtering, given its more general problem-solving features, has reached an important position in the field. The aim of this thesis is to design, implement and validate a hardware platform that constitutes itself an embedded intelligent system. The proposed system would combine particle filtering and evolutionary computation algorithms to generate intelligent behavior. Traditional approaches to particle filtering or evolutionary computation have been developed in software platforms, including parallel capabilities to some extent. In this work, an additional goal is fully exploiting hardware implementation advantages. By using the computational resources available in a FPGA device, better performance results in terms of computation time are expected. These hardware resources will be in charge of extensive repetitive computations. With this hardware-based implementation, real-time features are also expected.

Desenvolvimento de sistema automatizado de monitoramento de ambientes de produção animal, utilizando uma rede de sensores sem fio

As redes de sensores sem fio, aplicadas à automação do controle de ambientes representam um paradigma emergente da computação, onde múltiplos nós providos de sensores, sistemas computacionais autônomos e capacidade de comunicação sem fio, conformam uma rede cuja topologia altamente dinâmica permite adquirir informações sobre sistemas complexos sendo monitorados. Um dos fatores essenciais para obter um ganho na produtividade avícola é o controle da ambiência animal. Atualmente os métodos utilizados para o monitoramento e controle ambiental não podem considerar a grande quantidade de microambientes internos nos ambientes de produção animal e também requerem infraestruturas cabeadas complexas. Dentro desse contexto o objetivo deste trabalho foi desenvolver e testar um sistema automatizado de controle ambiental, através da utilização de sensores sem fio, que auxilie e proporcione maior segurança no controle de ambientes automatizados. O sistema monitora variáveis que influenciam na produtividade de aves, tais como temperatura e umidade e outras variáveis físico-químicas do aviário. A infraestrutura desenvolvida foi testada em um aviário experimental e resultou em um sistema seguro e com grande escalabilidade, que é capaz de controlar e monitorar o ambiente e ainda coletar e gravar dados. Foi utilizado o protocolo ZigBee® para gerenciar o fluxo de dados do sistema. Foram feitas análises da eficiência de comunicação do sistema no aviário, monitorando os pacotes de dados perdidos. Os testes demonstraram uma perda de dados de aproximadamente 2% dos pacotes enviados, demonstrando a eficiência das redes ZigBee® para gerenciar o fluxo de dados no interior do aviário. Desta forma, pode-se concluir que é possível e viável a implantação de uma rede ZigBee®, para automatizar ambientes de produção animal com coleta de dados em tempo real, utilizando um sistema integrado via internet, que compreende: instrumentação eletrônica, comunicação sem fio e engenharia de software\".

A robust and fast method for 6DoF motion estimation from generalized 3D data

Nowadays, there is an increasing number of robotic applications that need to act in real three-dimensional (3D) scenarios. In this paper we present a new mobile robotics orientated 3D registration method that improves previous Iterative Closest Points based solutions both in speed and accuracy. As an initial step, we perform a low cost computational method to obtain descriptions for 3D scenes planar surfaces. Then, from these descriptions we apply a force system in order to compute accurately and efficiently a six degrees of freedom egomotion. We describe the basis of our approach and demonstrate its validity with several experiments using different kinds of 3D sensors and different 3D real environments.

Active Visual Features Based on Events to Guide Robot Manipulators in Tracking Tasks

Traditional visual servoing systems do not deal with the topic of moving objects tracking. When these systems are employed to track a moving object, depending on the object velocity, visual features can go out of the image, causing the fail of the tracking task. This occurs specially when the object and the robot are both stopped and then the object starts the movement. In this work, we have employed a retina camera based on Address Event Representation (AER) in order to use events as input in the visual servoing system. The events launched by the camera indicate a pixel movement. Event visual information is processed only at the moment it occurs, reducing the response time of visual servoing systems when they are used to track moving objects.

Evaluating alternative gait strategies using evolutionary robotics

Evolutionary robitics is a branch of artificial intelligence concerned with the automatic generation of autonomous robots. Usually the form of the robit is predefined an various computational techniques are used to control the machine's behaviour. One aspect is the spontaneous generation of walking in legged robots and this can be used to investigate the mechanical requiements for efficient walking in bipeds. This paper demonstrates a bipedal simulator that spontaneously generates walking and running gaits. The model can be customized to represent a range of hominoid morphologies and used to predict performance paramets such as preferred speed and metabolic energy cost. Because it does not require any motion capture data it is particularly suitable for investigating locomotion in fossil animals. The predictoins for modern humans are highly accurate in terms of energy cost for a given speend and thus the values predicted for other bipeds are likely to be good estimates. To illustrate this the cost of transport is calculated for Australopithecus afarensis. The model allows the degree of maximum extension at the knee to be varied causing the model to adopt walking gaits varying from chimpanzee-like to human=like. The energy costs associated with these gait choices can thus be calculated and this information used to evaluate possible locomotor strategies in early hominids

OPA1-related disorders: Diversity of clinical expression, modes of inheritance and pathophysiology

International audience

Dynamic Bayesian network for semantic place classification in mobile robotics

In this paper, the problem of semantic place categorization in mobile robotics is addressed by considering a time-based probabilistic approach called dynamic Bayesian mixture model (DBMM), which is an improved variation of the dynamic Bayesian network. More specifically, multi-class semantic classification is performed by a DBMM composed of a mixture of heterogeneous base classifiers, using geometrical features computed from 2D laserscanner data, where the sensor is mounted on-board a moving robot operating indoors. Besides its capability to combine different probabilistic classifiers, the DBMM approach also incorporates time-based (dynamic) inferences in the form of previous class-conditional probabilities and priors. Extensive experiments were carried out on publicly available benchmark datasets, highlighting the influence of the number of time-slices and the effect of additive smoothing on the classification performance of the proposed approach. Reported results, under different scenarios and conditions, show the effectiveness and competitive performance of the DBMM.

Continual learning on nano drones

Miniaturized flying robotic platforms, called nano-drones, have the potential to revolutionize the autonomous robots industry sector thanks to their very small form factor. The nano-drones’ limited payload only allows for a sub-100mW microcontroller unit for the on-board computations. Therefore, traditional computer vision and control algorithms are too computationally expensive to be executed on board these palm-sized robots, and we are forced to rely on artificial intelligence to trade off accuracy in favor of lightweight pipelines for autonomous tasks. However, relying on deep learning exposes us to the problem of generalization since the deployment scenario of a convolutional neural network (CNN) is often composed by different visual cues and different features from those learned during training, leading to poor inference performances. Our objective is to develop and deploy and adaptation algorithm, based on the concept of latent replays, that would allow us to fine-tune a CNN to work in new and diverse deployment scenarios. To do so we start from an existing model for visual human pose estimation, called PULPFrontnet, which is used to identify the pose of a human subject in space through its 4 output variables, and we present the design of our novel adaptation algorithm, which features automatic data gathering and labeling and on-device deployment. We therefore showcase the ability of our algorithm to adapt PULP-Frontnet to new deployment scenarios, improving the R2 scores of the four network outputs, with respect to an unknown environment, from approximately [−0.2, 0.4, 0.0,−0.7] to [0.25, 0.45, 0.2, 0.1]. Finally we demonstrate how it is possible to fine-tune our neural network in real time (i.e., under 76 seconds), using the target parallel ultra-low power GAP 8 System-on-Chip on board the nano-drone, and we show how all adaptation operations can take place using less than 2mWh of energy, a small fraction of the available battery power.

Robots and humans as co-workers? The human-centred perspective of work with autonomous systems

The design of work organisation systems with automated equipment is facing new challenges and the emergence of new concepts. The social aspects that are related with new concepts on the complex work environments (CWE) are becoming more relevant for that design. The work with autonomous systems implies options in the design of workplaces. Especially that happens in such complex environments. The concepts of “agents”, “co-working” or “human-centred technical systems” reveal new dimensions related to human-computer interaction (HCI). With an increase in the number and complexity of those human-technology interfaces, the capacities of human intervention can become limited, originating further problems. The case of robotics is used to exemplify the issues related with automation in working environments and the emergence of new HCI approaches that would include social implications. We conclude that studies on technology assessment of industrial robotics and autonomous agents on manufacturing environment should also focus on the human involvement strategies in organisations. A needed participatory strategy implies a new approach to workplaces design. This means that the research focus must be on the relation between technology and social dimensions not as separate entities, but integrated in the design of an interaction system.

A Beacon based Localization System for Autonomous Mobile Robots and its Applications in Traffic and Transport Control

ACCURATE sensing of vehicle position and attitude is still a very challenging problem in many mobile robot applications. The mobile robot vehicle applications must have some means of estimating where they are and in which direction they are heading. Many existing indoor positioning systems are limited in workspace and robustness because they require clear lines-of-sight or do not provide absolute, driftfree measurements.The research work presented in this dissertation provides a new approach to position and attitude sensing system designed specifically to meet the challenges of operation in a realistic, cluttered indoor environment, such as that of an office building, hospital, industrial or warehouse. This is accomplished by an innovative assembly of infrared LED source that restricts the spreading of the light intensity distribution confined to a sheet of light and is encoded with localization and traffic information. This Digital Infrared Sheet of Light Beacon (DISLiB) developed for mobile robot is a high resolution absolute localization system which is simple, fast, accurate and robust, without much of computational burden or significant processing. Most of the available beacon's performance in corridors and narrow passages are not satisfactory, whereas the performance of DISLiB is very encouraging in such situations. This research overcomes most of the inherent limitations of existing systems.The work further examines the odometric localization errors caused by over count readings of an optical encoder based odometric system in a mobile robot due to wheel-slippage and terrain irregularities. A simple and efficient method is investigated and realized using an FPGA for reducing the errors. The detection and correction is based on redundant encoder measurements. The method suggested relies on the fact that the wheel slippage or terrain irregularities cause more count readings from the encoder than what corresponds to the actual distance travelled by the vehicle.The application of encoded Digital Infrared Sheet of Light Beacon (DISLiB) system can be extended to intelligent control of the public transportation system. The system is capable of receiving traffic status input through a GSM (Global System Mobile) modem. The vehicles have infrared receivers and processors capable of decoding the information, and generating the audio and video messages to assist the driver. The thesis further examines the usefulness of the technique to assist the movement of differently-able (blind) persons in indoor or outdoor premises of his residence.The work addressed in this thesis suggests a new way forward in the development of autonomous robotics and guidance systems. However, this work can be easily extended to many other challenging domains, as well.

A Floating-point Extended Kalman Filter Implementation for Autonomous Mobile Robots

Localization and Mapping are two of the most important capabilities for autonomous mobile robots and have been receiving considerable attention from the scientific computing community over the last 10 years. One of the most efficient methods to address these problems is based on the use of the Extended Kalman Filter (EKF). The EKF simultaneously estimates a model of the environment (map) and the position of the robot based on odometric and exteroceptive sensor information. As this algorithm demands a considerable amount of computation, it is usually executed on high end PCs coupled to the robot. In this work we present an FPGA-based architecture for the EKF algorithm that is capable of processing two-dimensional maps containing up to 1.8 k features at real time (14 Hz), a three-fold improvement over a Pentium M 1.6 GHz, and a 13-fold improvement over an ARM920T 200 MHz. The proposed architecture also consumes only 1.3% of the Pentium and 12.3% of the ARM energy per feature.

Autonomous mobile robots designing for the medical trash collector task

This paper describes the UNESP robotic team in the medical trash collector task, proposed on the 5 rd IEEE Latin American Robots Competition in the LEGO category. We present our understanding of the task and discuss the proposed solution, focusing on the mechanical and computational issues of the robots. The mechanics is based on rigid body capability of transforming rotational into curvilinear movement. With respect to the computational control, the system is modeled as a reactive system with sequential transition of behaviors. A state-machine is proposed to allow this transition, and the synchronization of robotic states is guaranteed by the communication system. The proposed approach has shown itself capable of dealing with the high difficulty degree of this cooperative task. ©2006 IEEE.

Petri nets and autonomous systems: A case study

This work presents the Petri net-based modeling of an autonomous robot's navigation system used for the application of supplies in agriculture. The model was developed theoretically and implemented through the CPNTools software. It simulates the behavior of the robot, capturing environmental characteristics by means of sensors, making appropriate decisions, and forwarding them to the corresponding actuators. By exciting the model using CPNTools it is possible to simulate situations that the robot might undergo, without the need to expose it to real potentially dangerous situations. ©2009 IEEE.

TAO - A software platform for autonomous mobile robots

To date, different techniques of navigation for mobile robots have been developed. However, the experimentation of these techniques is not a trivial task because usually it is not possible to reuse the developed control software due to system incompabilities. This paper proposes a software platform that provides means for creating reusable software modules through the standardization of software interfaces, which represent the various robot modules. © 2012 ICROS.

Coordination and Control of Autonomous Mobile Robots Swarms by using Particle Swarm Optimization Algorithm and Consensus Theory

This thesis presents some different techniques designed to drive a swarm of robots in an a-priori unknown environment in order to move the group from a starting area to a final one avoiding obstacles. The presented techniques are based on two different theories used alone or in combination: Swarm Intelligence (SI) and Graph Theory. Both theories are based on the study of interactions between different entities (also called agents or units) in Multi- Agent Systems (MAS). The first one belongs to the Artificial Intelligence context and the second one to the Distributed Systems context. These theories, each one from its own point of view, exploit the emergent behaviour that comes from the interactive work of the entities, in order to achieve a common goal. The features of flexibility and adaptability of the swarm have been exploited with the aim to overcome and to minimize difficulties and problems that can affect one or more units of the group, having minimal impact to the whole group and to the common main target. Another aim of this work is to show the importance of the information shared between the units of the group, such as the communication topology, because it helps to maintain the environmental information, detected by each single agent, updated among the swarm. Swarm Intelligence has been applied to the presented technique, through the Particle Swarm Optimization algorithm (PSO), taking advantage of its features as a navigation system. The Graph Theory has been applied by exploiting Consensus and the application of the agreement protocol with the aim to maintain the units in a desired and controlled formation. This approach has been followed in order to conserve the power of PSO and to control part of its random behaviour with a distributed control algorithm like Consensus.