Control and navigation of a rover for agricultural purposes

L’obiettivo di questa tesi è di descrivere e implementare via software un modello di rover autonomo per uso in ambito agricolo. La scelta di questo argomento deriva dal fatto che al laboratorio CASY dell’Università di Bologna è stato commissionato un robot che possa aiutare piccoli imprenditori agricoli a essere competitivi con i più grandi. Le funzionalità che il robot avrà, una volta ultimato, andranno dal tagliare l’erba allo spruzzare fertilizzante sugli alberi da frutto. Questa tesi si interessa del progetto del sistema di navigazione. Inizialmente viene introdotto il modello cinematico e in particolare la configurazione differential drive in cui il rover rientra. Successivamente viene elaborato un sistema di controllo basato sulla linearizzazione statica del feedback. Una volta completati il modello e il sistema di controllo si procede con la generazione di traiettoria: vengono analizzati e confrontati alcuni algoritmi per l’inseguimento di una traiettoria definita tramite waypoint. Infine è presentato un algoritmo per la navigazione all’interno di un campo di filari di alberi da frutto. Le uniche informazioni esterne disponibili in questo contesto sono le rilevazioni di sensori di distanza frontali e laterali, in quanto un GPS sarebbe troppo impreciso per gli scopi. Questa tesi costituisce la base per ulteriori sviluppi del progetto. In particolare la realizzazione di un programma di supervisione che stabilisca la modalità di moto da attuare e programmi specifici per le varie funzionalità agricole del rover.

FROG (formulation of a robot organization that generalizes) : a hardwardware [sic] experiment in machine cognition /

Vita.

Application of Fractional Calculus in the Dynamical Analysis and Control of Mechanical Manipulators

Mathematics Subject Classification: 26A33, 93C83, 93C85, 68T40

Design, development and control of a new generation high performance linear actuator for parallel robots and other applications

The main focus of this research is to design and develop a high performance linear actuator based on a four bar mechanism. The present work includes the detailed analysis (kinematics and dynamics), design, implementation and experimental validation of the newly designed actuator. High performance is characterized by the acceleration of the actuator end effector. The principle of the newly designed actuator is to network the four bar rhombus configuration (where some bars are extended to form an X shape) to attain high acceleration. Firstly, a detailed kinematic analysis of the actuator is presented and kinematic performance is evaluated through MATLAB simulations. A dynamic equation of the actuator is achieved by using the Lagrangian dynamic formulation. A SIMULINK control model of the actuator is developed using the dynamic equation. In addition, Bond Graph methodology is presented for the dynamic simulation. The Bond Graph model comprises individual component modeling of the actuator along with control. Required torque was simulated using the Bond Graph model. Results indicate that, high acceleration (around 20g) can be achieved with modest (3 N-m or less) torque input. A practical prototype of the actuator is designed using SOLIDWORKS and then produced to verify the proof of concept. The design goal was to achieve the peak acceleration of more than 10g at the middle point of the travel length, when the end effector travels the stroke length (around 1 m). The actuator is primarily designed to operate in standalone condition and later to use it in the 3RPR parallel robot. A DC motor is used to operate the actuator. A quadrature encoder is attached with the DC motor to control the end effector. The associated control scheme of the actuator is analyzed and integrated with the physical prototype. From standalone experimentation of the actuator, around 17g acceleration was achieved by the end effector (stroke length was 0.2m to 0.78m). Results indicate that the developed dynamic model results are in good agreement. Finally, a Design of Experiment (DOE) based statistical approach is also introduced to identify the parametric combination that yields the greatest performance. Data are collected by using the Bond Graph model. This approach is helpful in designing the actuator without much complexity.

Localization of a mobile robot based on odometry and natural landmarks using extended kalman filter

SANTANA, André M.; SOUZA, Anderson A. S.; BRITTO, Ricardo S.; ALSINA, Pablo J.; MEDEIROS, Adelardo A. D. Localization of a mobile robot based on odometry and natural landmarks using extended Kalman Filter. In: INTERNATIONAL CONFERENCE ON INFORMATICS IN CONTROL, AUTOMATION AND ROBOTICS, 5., 2008, Funchal, Portugal. Proceedings... Funchal, Portugal: ICINCO, 2008.

Shape-based compliance control for snake robots

I serpenti robot sono una classe di meccanismi iper-ridondanti che appartiene alla robotica modulare. Grazie alla loro forma snella ed allungata e all'alto grado di ridondanza possono muoversi in ambienti complessi con elevata agilità. L'abilità di spostarsi, manipolare e adattarsi efficientemente ad una grande varietà di terreni li rende ideali per diverse applicazioni, come ad esempio attività di ricerca e soccorso, ispezione o ricognizione. I robot serpenti si muovono nello spazio modificando la propria forma, senza necessità di ulteriori dispositivi quali ruote od arti. Tali deformazioni, che consistono in movimenti ondulatori ciclici che generano uno spostamento dell'intero meccanismo, vengono definiti andature. La maggior parte di esse sono ispirate al mondo naturale, come lo strisciamento, il movimento laterale o il movimento a concertina, mentre altre sono create per applicazioni specifiche, come il rotolamento o l'arrampicamento. Un serpente robot con molti gradi di libertà deve essere capace di coordinare i propri giunti e reagire ad ostacoli in tempo reale per riuscire a muoversi efficacemente in ambienti complessi o non strutturati. Inoltre, aumentare la semplicità e ridurre il numero di controllori necessari alla locomozione alleggerise una struttura di controllo che potrebbe richiedere complessità per ulteriori attività specifiche. L'obiettivo di questa tesi è ottenere un comportamento autonomo cedevole che si adatti alla conformazione dell'ambiente in cui il robot si sta spostando, accrescendo le capacità di locomozione del serpente robot. Sfruttando la cedevolezza intrinseca del serpente robot utilizzato in questo lavoro, il SEA Snake, e utilizzando un controllo che combina cedevolezza attiva ad una struttura di coordinazione che ammette una decentralizzazione variabile del robot, si dimostra come tre andature possano essere modificate per ottenere una locomozione efficiente in ambienti complessi non noti a priori o non modellabili.

Développement et expérimentation d'algorithmes de réorientation pour un robot sériel en chute libre

Ce mémoire présente 2 types de méthodes pour effectuer la réorientation d’un robot sériel en chute libre en utilisant les mouvements internes de celui-ci. Ces mouvements sont prescrits à partir d’algorithmes de planification de trajectoire basés sur le modèle dynamique du robot. La première méthode tente de réorienter le robot en appliquant une technique d’optimisation locale fonctionnant avec une fonction potentielle décrivant l’orientation du système, et la deuxième méthode applique des fonctions sinusoïdales aux articulations pour réorienter le robot. Pour tester les performances des méthodes en simulation, on tente de réorienter le robot pour une configuration initiale et finale identiques où toutes les membrures sont alignées mais avec le robot ayant complété une rotation de 180 degrés sur lui-même. Afin de comparer les résultats obtenus avec la réalité, un prototype de robot sériel plan flottant possédant trois membrures et deux liaisons rotoïdes est construit. Les expérimentations effectuées montrent que le prototype est capable d’atteindre les réorientations prescrites si peu de perturbations extérieures sont présentes et ce, même si le contrôle de l’orientation est effectué en boucle ouverte.

Multirobot Tethering for Localization and Control

Particle filtering has proven to be an effective localization method for wheeled autonomous vehicles. For a given map, a sensor model, and observations, occasions arise where the vehicle could equally likely be in many locations of the map. Because particle filtering algorithms may generate low confidence pose estimates under these conditions, more robust localization strategies are required to produce reliable pose estimates. This becomes more critical if the state estimate is an integral part of system control. We investigate the use of particle filter estimation techniques on a hovercraft vehicle. The marginally stable dynamics of a hovercraft require reliable state estimates for proper stability and control. We use the Monte Carlo localization method, which implements a particle filter in a recursive state estimate algorithm. An H-infinity controller, designed to accommodate the latency inherent in our state estimation, provides stability and controllability to the hovercraft. In order to eliminate the low confidence estimates produced in certain environments, a multirobot system is designed to introduce mobile environment features. By tracking and controlling the secondary robot, we can position the mobile feature throughout the environment to ensure a high confidence estimate, thus maintaining stability in the system. A laser rangefinder is the sensor the hovercraft uses to track the secondary robot, observe the environment, and facilitate successful localization and stability in motion.

Localization of a mobile robot based on odometry and natural landmarks using extended kalman filter

SANTANA, André M.; SOUZA, Anderson A. S.; BRITTO, Ricardo S.; ALSINA, Pablo J.; MEDEIROS, Adelardo A. D. Localization of a mobile robot based on odometry and natural landmarks using extended Kalman Filter. In: INTERNATIONAL CONFERENCE ON INFORMATICS IN CONTROL, AUTOMATION AND ROBOTICS, 5., 2008, Funchal, Portugal. Proceedings... Funchal, Portugal: ICINCO, 2008.

ENABLING HARDWARE TECHNOLOGIES FOR AUTONOMY IN TINY ROBOTS: CONTROL, INTEGRATION, ACTUATION

The last two decades have seen many exciting examples of tiny robots from a few cm3 to less than one cm3. Although individually limited, a large group of these robots has the potential to work cooperatively and accomplish complex tasks. Two examples from nature that exhibit this type of cooperation are ant and bee colonies. They have the potential to assist in applications like search and rescue, military scouting, infrastructure and equipment monitoring, nano-manufacture, and possibly medicine. Most of these applications require the high level of autonomy that has been demonstrated by large robotic platforms, such as the iRobot and Honda ASIMO. However, when robot size shrinks down, current approaches to achieve the necessary functions are no longer valid. This work focused on challenges associated with the electronics and fabrication. We addressed three major technical hurdles inherent to current approaches: 1) difficulty of compact integration; 2) need for real-time and power-efficient computations; 3) unavailability of commercial tiny actuators and motion mechanisms. The aim of this work was to provide enabling hardware technologies to achieve autonomy in tiny robots. We proposed a decentralized application-specific integrated circuit (ASIC) where each component is responsible for its own operation and autonomy to the greatest extent possible. The ASIC consists of electronics modules for the fundamental functions required to fulfill the desired autonomy: actuation, control, power supply, and sensing. The actuators and mechanisms could potentially be post-fabricated on the ASIC directly. This design makes for a modular architecture. The following components were shown to work in physical implementations or simulations: 1) a tunable motion controller for ultralow frequency actuation; 2) a nonvolatile memory and programming circuit to achieve automatic and one-time programming; 3) a high-voltage circuit with the highest reported breakdown voltage in standard 0.5 μm CMOS; 4) thermal actuators fabricated using CMOS compatible process; 5) a low-power mixed-signal computational architecture for robotic dynamics simulator; 6) a frequency-boost technique to achieve low jitter in ring oscillators. These contributions will be generally enabling for other systems with strict size and power constraints such as wireless sensor nodes.

Integración de Arbotix, Raspberry Pi y motores Dynamixel Ax-12+ para un robot humanoide que busca y patea pelotas

En este artículo se presenta a DeBuPa (Detección Búsqueda Pateo) un humanoide de tamaño pequeño (38 cm de alto) construido con las piezas del kit Bioloid. Del kit se ha excluido la tarjeta CM-510 para sustituirla por la tarjeta controladora Arbotix, que será la que controle los 16 motores Dynamixel Ax-12+ (para mover al robot) y 2 servomotores analógicos (para mover la cámara). Además se ha agregado un mini computador Raspberry Pi, con su cámara, para que el robot pueda detectar y seguir la pelota de forma autónoma. Todos estos componentes deben ser coordinados para que se logre cumplir la tarea de detectar, seguir y patear la pelota. Por ello se hace necesaria la comunicación entre la Arbotix y la Raspberry Pi. La herramienta empleada para ello es el framework ROS (Robot Operating System). En la Raspberry Pi se usa el lenguaje C++ y se ejecuta un solo programa encargado de captar la imagen de la cámara, filtrar y procesar para encontrar la pelota, tomar la decisión de la acción a ejecutar y hacer la petición a la Arbotix para que dé la orden a los motores de ejecutar el movimiento. Para captar la imagen de la cámara se ha utilizado la librería RasPiCam CV. Para filtrar y procesar la imagen se ha usado las librerías de OpenCV. La Arbotix, además de controlar los motores, se encarga de monitorizar que el robot se encuentre balanceado, para ello usa el sensor Gyro de Robotis. Si detecta un desbalance de un cierto tamaño puede saber si se ha caído y levantarse.

Mimicking the behaviour of idiotypic AIS robot controllers using probabilistic systems

Previous work has shown that robot navigation systems that employ an architecture based upon the idiotypic network theory of the immune system have an advantage over control techniques that rely on reinforcement learning only. This is thought to be a result of intelligent behaviour selection on the part of the idiotypic robot. In this paper an attempt is made to imitate idiotypic dynamics by creating controllers that use reinforcement with a number of different probabilistic schemes to select robot behaviour. The aims are to show that the idiotypic system is not merely performing some kind of periodic random behaviour selection, and to try to gain further insight into the processes that govern the idiotypic mechanism. Trials are carried out using simulated Pioneer robots that undertake navigation exercises. Results show that a scheme that boosts the probability of selecting highly-ranked alternative behaviours to 50% during stall conditions comes closest to achieving the properties of the idiotypic system, but remains unable to match it in terms of all round performance.

Control para sistemas LPV basado en la parametrización de controladores estabilizantes

Este trabajo se enfoca en el estudio del control de sistemas Multi-Entrada Multi-Salida (MIMO) Lineales con Parámetros Variantes en el Tiempo (LPV). Los parámetros son medibles y permanecen dentro de cotas conocidas. El control por retroalimentación de salida garantiza estabilidad cuadrática (QS) y desempeño, mediante el Teorema de los vértices y el Lema de Cota Real (BRL). Se proponen condiciones para que el sistema retroalimentado sea convexo cuando se utilizan controladores estabilizantes en cada vértice. El controlador LPV resulta de la interpolación de estos controladores, y se estudia la relación entre la estabilidad y el desempeño del control de los vértices, y la estabilidad y desempeño del sistema LPV. Además, se da una forma explícita del parámetro libre de la Parametrización de Todos los Controladores Estabilizantes (PTCE) que resuelve un criterio de sensibilidad mezclada cuando se tiene un modelo de incertidumbre aditivo a la salida. Los resultados se aplican a un robot planar rotacional de dos grados de libertad, a un motor de CD y a un sistema de dos masas.