Animals are often able to solve complex navigational tasks in very challenging terrain, despite using low resolution sensors and minimal computational power, providing inspiration for robots. In particular, many species of insect are known to solve complex navigation problems, often combining an array of different behaviours (Wehner et al., 1996; Collett, 1996). Their nervous system is also comparatively simple, relative to that of mammals and other vertebrates. In the first part of this thesis, the visual input of a navigating desert ant, Cataglyphis velox, was mimicked by capturing images in ultraviolet (UV) at similar wavelengths to the ant’s compound eye. The natural segmentation of ground and sky lead to the hypothesis that skyline contours could be used by ants as features for navigation. As proof of concept, sky-segmented binary images were used as input for an established localisation algorithm SeqSLAM (Milford and Wyeth, 2012), validating the plausibility of this claim (Stone et al., 2014). A follow-up investigation sought to determine whether using the sky as a feature would help overcome image matching problems that the ant often faced, such as variance in tilt and yaw rotation. A robotic localisation study showed that using spherical harmonics (SH), a representation in the frequency domain, combined with extracted sky can greatly help robots localise on uneven terrain. Results showed improved performance to state of the art point feature localisation methods on fast bumpy tracks (Stone et al., 2016a). In the second part, an approach to understand how insects perform a navigational task called path integration was attempted by modelling part of the brain of the sweat bee Megalopta genalis. A recent discovery that two populations of cells act as a celestial compass and visual odometer, respectively, led to the hypothesis that circuitry at their point of convergence in the central complex (CX) could give rise to path integration. A firing rate-based model was developed with connectivity derived from the overlap of observed neural arborisations of individual cells and successfully used to build up a home vector and steer an agent back to the nest (Stone et al., 2016b). This approach has the appeal that neural circuitry is highly conserved across insects, so findings here could have wide implications for insect navigation in general. The developed model is the first functioning path integrator that is based on individual cellular connections.

No description available.

Dans ce travail, nous avons premièrement développé et caractérisé des capteurs de flux optique robustes aux changements de conditions lumineuses inspirés par le système visuel de la mouche et mesurant la vitesse angulaire à l'aide de l'algorithme appelé "time of travel". En particulier, nous avons comparé les performances de capteurs mesurant visuellement la vitesse angulaire en intérieur et en extérieur. Les résultats de nos capteurs bio-inspirés ont aussi été comparés avec des capteurs de souris optique. Enfin, une nouvelle implémentation de l'algorithme "time of travel" a été proposée réduisant la charge de calcul de l'unité de traitement.Dans le cadre du projet européen CurvACE (Curved Artificial Compound Eye), nous avons aussi participé au développement du premier oeil composé courbé artificiel capable de mesurer le flux optique à haute vitesse sur une large gamme de lumière ambiante. En particulier, nous avons caractérisé ce capteur et montré sa capacité à mesurer le flux optique à l'aide de plusieurs algorithmes.Finalement, nous avons aussi développé un robot aérien miniature attaché appelé BeeRotor équipé de capteurs et de stratégies de vol imitant les insectes volants et se déplaçant de manière autonome dans un tunnel contrasté. Ce robot peut expliquer comment les abeilles contrôlent leur vitesse et leur position à l'aide du flux optique, tout en démontrant que des solutions alternatives existent aux systèmes couramment utilisés en robotique. Basé seulement sur des boucles de contrôle réagissant à l'environnement, cet hélicoptère a démontré sa capacité à voler de manière autonome dans un environnement complexe et mobile. / In this thesis, we first developed and characterized optic flow sensors robust to illuminance changes inspired by the visual system of the fly and computing the angular speed thanks to the "time of travel" scheme. In particular, we have compared the performances of sensors processing the visual angular speed based on a standard retina or an aVLSI retina composed of pixels automatically adapting to the background illuminance in indoor and outdoor environments. The results of such bio-inspired sensors have also been compared with optic mouse sensors which are used nowadays on Micro Aerial Vehicles to process the optic flow but only in outdoor environments. Finally, a new implementation of the "time of travel" scheme has been proposed reducing the computational load of the processing unit.In the framework of the European project CurvACE, we also participated to the design and development of the first curved artificial compound eye including fast motion detection in a very large range of illuminations. In particular, we characterized such sensor showing its ability to extract optic flow using different algorithms.Finally, we also developed a tethered miniature aerial robot equipped with sensors and control strategies mimicking flying insects navigating in a high-roof tunnel. This robot may explain how honeybees control their speed and position thanks to optic flow, while demonstrating alternative solution to classical robotic approach relying on ground-truth and metric sensors. Based only on visuomotor control loops reacting suitably to the environment, this rotorcraft has shown its ability to fly autonomously in complex and unstationary tunnels.

No description available.

Musculoskeletal disorders, specifically low back pain, has been well documented andreported by astronauts throughout the space exploration era. Statistics from astronautmemoirs states that 52-68% of astronauts experience moderate to severe lower backpain after prolonged spaceflights. The main cause is atrophy in the paraspinal musclesof the lumbar region. No sufficient countermeasure exists in-flight currently and therehabilitation programs remain ineffective. This thesis presents the first attempt to designand develop a prototype robotic exoskeleton, actuated by pneumatic artificial muscles, asan active countermeasure in-flight where it will be utilised as an equipment for muscularhypertrophy and a supporting device for rehabilitation programs on Earth. It will bemanufactured by additive manufacturing methods for adaptability while remaining lowin weight.A thorough analysis of the spine and lumbar region as a biomechanical system wasmade. Appropriate assumptions was made to simplify the understanding of the complexsystem that is the human spine. The targeted muscles were: multifidus, erector spinaeiliocostalis and erector spinae longissimus. A force analysis of the human torso bendingin the sagittal plane was made, finding that the torques of the torso reaches 244 Nm.The complete exoskeleton design is presented with the parts that will be 3D-printed andthe working principle of the system. Thereafter an extensive model of the exoskeletonis established using Denavit-Hartenberg representation of manipulators as a serial linksystem. The model provides a fundamental understanding of exoskeleton and enablesthe possibility to simulate it accurately. The evaluation protocol for the validation testsis then presented. Active pressure will be tested at 0, 3 and 6 bar and loads of 5 and 11kg will be lifted.Subsequently the assembly, with all the hardware and software selected for the prototypeis demonstrated. Thereafter the results of the evaluation tests are presented followed bya discussion of the results; anomalies, faults and challenges are subjects discussed. Thediscussion concludes that the exoskeleton shows potential for both supporting the motionin a rehabilitation use and enabling muscular hypertrophy in the lumbar region for theresistive tests. Although an extensive heavy-duty evaluation needs to be performed totruly validate the exoskeleton.

Ideal prostheses are defined as artificial limbs that would permit physically impaired individuals freedom of movement and independence rather than a life of disability and dependence. Current lower limb prostheses range from a single mechanical revolute joint to advanced microprocessor controlled mechanisms. Despite the advancement in technology and medicine, current lower limb prostheses are still lacking an actuation element, which prohibits patients from regaining their original mobility and improving their quality of life. This thesis aims to design and test a Pneumatic Artificial Muscle that would actuate lower limb prostheses. This would offer patients the ability to ascend and descend stairs as well as standing up from a sitting position. A comprehensive study of knee biomechanics is first accomplished to characterize the actuation requirement, and subsequently a Pneumatic Artificial Muscle design is proposed. A novel design of muscle end fixtures is presented which would allow the muscle to operate at a gage pressure surpassing 2.76 MPa (i.e. 400 psi) and yield a muscle force that is at least 3 times greater than that produced by any existing equivalent Pneumatic Artificial Muscle. Finally, the proposed Pneumatic Artificial Muscle is tested and validated to verify that it meets the size, weight, kinetic and kinematic requirements of human knee articulation.

Ideal prostheses are defined as artificial limbs that would permit physically impaired individuals freedom of movement and independence rather than a life of disability and dependence. Current lower limb prostheses range from a single mechanical revolute joint to advanced microprocessor controlled mechanisms. Despite the advancement in technology and medicine, current lower limb prostheses are still lacking an actuation element, which prohibits patients from regaining their original mobility and improving their quality of life. This thesis aims to design and test a Pneumatic Artificial Muscle that would actuate lower limb prostheses. This would offer patients the ability to ascend and descend stairs as well as standing up from a sitting position. A comprehensive study of knee biomechanics is first accomplished to characterize the actuation requirement, and subsequently a Pneumatic Artificial Muscle design is proposed. A novel design of muscle end fixtures is presented which would allow the muscle to operate at a gage pressure surpassing 2.76 MPa (i.e. 400 psi) and yield a muscle force that is at least 3 times greater than that produced by any existing equivalent Pneumatic Artificial Muscle. Finally, the proposed Pneumatic Artificial Muscle is tested and validated to verify that it meets the size, weight, kinetic and kinematic requirements of human knee articulation.