Insect-Inspired Visual Perception for Flight Control and Collision Avoidance

Flying robots are increasingly used for tasks such as aerial mapping, fast exploration, video footage and monitoring of buildings. Autonomous flight at low altitude in cluttered and unknown environments is an active research topic because it poses challenging perception and control problems. Traditional methods for collision-free navigation at low altitude require heavy resources to deal with the complexity of natural environments, something that limits the autonomy and the payload of flying robots.

Flying insects, however, are able to navigate safely and efficiently using vision as the main sensory modality. Flying insects rely on low resolution, high refresh rate, and wide-angle compound eyes to extract angular image motion and move in unstructured environments. These strategies result in systems that are physically and computationally lighter than those often found in high-definition stereovision.

Taking inspiration from insects offers great potential for building small flying robots capable of navigating in cluttered environments using lightweight vision sensors. In this thesis, we investigate insect perception of visual motion and insect vision based flight control in cluttered environments. We use the knowledge gained through the modelling of neural circuits and behavioural experiments to develop flying robots with insect-inspired control strategies for goal-oriented navigation in complex environments.

We start by exploring insect perception of visual motion. We present a study that reconciles an apparent contradiction in the literature for insect visual control: current models developed to explain insect flight behaviour rely on the measurement of optic flow, however the most prominent neural model for visual motion extraction (the Elementary Motion Detector, or EMD) does not measure optic flow. We propose a model for unbiased optic flow estimation that relies on comparing the output of multiple EMDs pointed in varying viewing directions. Our model is of interest of both engineers and biologists because it is computationally more efficient than other optic flow estimation algorithms, and because it represents a biologically plausible model for optic flow extraction in insect neural systems.

We then focus on insect flight control strategies in the presence of obstacles. By recording the trajectories of bumblebees (Bombus terrestris), and by comparing them to simulated flights, we show that bumblebees rely primarily on the frontal part of their field of view, and that they pool optic flow in two different manners for the control of flight speed and of lateral position. For the control of lateral position, our results suggest that bumblebees selectively react to the portions of the visual field where optic flow is the highest, which correspond to the closest obstacles.

Finally, we tackle goal-oriented navigation with a novel algorithm that combines aspects of insect perception and flight control presented in this thesis -- like the detection of fastest moving objects in the frontal visual field -- with other aspects of insect flight known from the literature such as saccadic flight pattern. Through simulations, we demonstrate autonomous navigation in forest-like environments using only local optic flow information and assuming knowledge about the direction to the navigation goal.

Enactive robot vision

Mototaka Suzuki

The complexity of today's autonomous robots poses a major challenge for Artificial Intelligence. These robots are equipped with sophisticated sensors and mechanical abilities that allow them to enter our homes and interact with humans. For example, today's robots are almost all equipped with vision and several of them can move over rough terrain with wheels or legs. The methods developed so far in Artificial Intelligence, however, are not yet ready to cope with the complexity of the information gathered through the robot sensors and the need for rapid action in partially unknown and dynamic environments. In this thesis, I will argue that the apparent complexity of the environment and of the robot brain can be significantly simplified if perception, behavior, and learning are allowed to co-develop on the same time scale. In doing so, robots become sensitive to, and actively exploit, characteristics of the environment that they can tackle within their own computational and physical constraints. This line of work is grounded on philosophical and psychological research showing that perception is an active process mediated by behavior. However, computational models of active vision are very rare and often rely on architectures that are preprogrammed to detect certain characteristics of the environment. Previous work have shown that complex visual tasks, such as position and size invariant shape recognition as well as navigation, can be tackled with remarkably simple neural architectures generated by a coevolutionary process of active vision and feature selection. Behavioral machines equipped with primitive vision systems and direct pathways between visual and motor neurons were evolved while freely interacting with their environments. I proceed further on this line of investigation and describe the application of this methodology in three situations, namely car driving with an omnidirectional camera, goal-oriented navigation of a humanoid robot, and cooperative tasks by two agents. I will show that these systems develop sensitivity to a number of oriented, retinotopic, visual features – oriented edges, corners, height – and a behavioral repertoire to locate, bring, and keep these features in sensitive regions of the vision system that allow them to accomplish their goals. In a second set of experiments, I will show that active vision can be exploited by the robot to perform anticipatory exploration of the environment in a task that requires landmark-based navigation. Evolved robots exploit an internal expectation system that makes use of active exploration to check for expected events in the environment. I will describe a third set of experiments where, in addition to an evolutionary process, the visual system of the robot can develop receptive fields by means of unsupervised Hebbian learning and show that these receptive fields are significantly affected by the behavior of the system and differ from those predicted by most computational models of visual cortex. Finally, I will show that these robots replicate the performance deficiencies observed in experiments of motor deprivation with kitten when they are exposed to the same type of motor deprivations. Furthermore, the analyses of our robot brains suggest an explanation for the deficiencies observed in kitten that have not yet been fully understood.

EPFL2007

Development of a robotic mobile mapping system by vision-aided inertial navigation

Fadi Atef Bayoud

Vision-based inertial-aided navigation is gaining ground due to its many potential applications. In previous decades, the integration of vision and inertial sensors was monopolised by the defence industry due to its complexity and unrealistic economic burden. After the technology advancement, high-quality hardware and computing power became reachable for the investigation and realisation of various applications. In this thesis, a mapping system by vision-aided inertial navigation was developed for areas where GNSS signals are unreachable, for example, indoors, tunnels, city canyons, forests, etc. In this framework, a methodology on the integration of vision and inertial sensors was presented, analysed and tested when the only available information at the beginning is a number of features with known location/coordinates (with no GNSS signals accessibility), thus employing the method of "SLAM: Simultaneous Localisation And Mapping". SLAM is a term used in the robotics community to describe the problem of mapping the environment and at the same time using this map to determine (or to help in determining) the location of the mapping device. In addition to this, a link between the robotics and geomatics community was established where briefly the similarities and differences were outlined in terms of handling the navigation and mapping problem. Albeit many differences, the goal is common: developing a "navigation and mapping system" that is not bounded to the limits imposed by the used sensors. Classically, terrestrial robotics SLAM is approached using LASER scanners to locate the robot relative to a structured environment and to map this environment at the same time. However, outdoors robotics SLAM is not feasible with LASER scanners alone due to the environment's roughness and absence of simple geometric features. Recently in the robotics community, the use of visual methods, integrated with inertial sensors, has gained an interest. These visual methods rely on one or more cameras (or video) and make use of a single Kalman Filter with a state vector containing the map and the robot coordinates. This concept introduces high non-linearity and complications to the filter, which then needs to run at high rates (more than 20 Hz) with simplified navigation and mapping models. In this study, SLAM is developed using the Geomatics Engineering approach. Two filters are used in parallel: the Least-Squares Adjustment (LSA) for feature coordinates determination and the Kalman Filter (KF) for navigation correction. For this, a mobile mapping system (independent of GPS) is introduced by employing two CCD cameras (one metre apart) and one IMU. Conceptually, the outputs of the LSA photogrammetric resection (position and orientation) are used as the external measurements for the inertial KF. The filtered position and orientation are subsequently employed in the Photogrammetric intersection to map the surrounding features that are used as control points for the resection in the next epoch. In this manner, the KF takes the form of a navigation only filter, with a state vector containing the corrections to the navigation parameters. This way, the mapping and localisation can be updated at low rates (1 to 2 Hz) and use more complete modelling. Results show that this method is feasible with limitation induced from the quality of the images and the number of used features. Although simulation showed that (depending on the image geometry) determining the features' coordinates with an accuracy of 5-10 cm for objects at distances of up to 10 metres is possible, in practice this is not achieved with the employed hardware and pixel measurement techniques. Navigational accuracies depend as well on the quality of the images and the number and accuracy of the points used in the resection. While more than 25 points are needed to achieve centimetre accuracy from resection, they have to be within a distance of 10 metres from the cameras; otherwise, the resulting resection output will be of insufficient accuracy and further integration quality deteriorates. The initial conditions highly affect SLAM performance; these are the method of IMU initialisation and the a-priori assumptions on error distribution. The geometry of the system will furthermore have a consequence on possible applications. To conclude, the development consisted in establishing a mathematical framework, as well as implementing methods and algorithms for a novel integration methodology between vision and inertial sensors. The implementation and validation of the software have presented the main challenges, and it can be considered the first of a kind where all components were developed from scratch, with no pre-existing modules. Finally, simulations and practical tests were carried out, from which initial conclusions and recommendations were drawn to build upon. It is the author's hope that this work will stimulate others to investigate further this interesting problem taking into account the conclusions and recommendations sketched herein.

EPFL2006

Miniature Cameras for Ultra-Light Flying Robots

Adam Klaptocz

The Laboratory of Intelligent Systems (LIS) at the Swiss Federal Technical Institute at Lausanne (EPFL) is working on a project to create an autonomous flying robot that uses ‘vision’ (i.e. optical flow) to maneuver through small spaces such as corridors in a building. The navigation algorithm uses data from a linear photodiode array and a gyroscope to determine the distance of the robot from an obstacle. This project involves the design of a light-weight vision system to be used on the flying robot. The vision system is made up of two identical modules, each one containing a linear photodiode sensor array, a focusing lens with associated plastic casing, a gyroscope, and the PCB on which the system is mounted. The first module will be pointed ahead of the robot and will be used for lateral steering, and the second module will be pointed straight down and be used to control elevation. The main design constraint is the weight of the system, which has to be below 2g for the robot to fly.