Robot path planning and dynamic obstacle avoidance are defined as a problem that robots plan a feasible path from a given starting point to a destination point in a nonlinear dynamic environment, and safely bypass dynamic obstacles to the destination with minimal deviation from the trajectory. Path planning is a typical sequential decision-making problem. Dynamic local observable environment requires real-time and adaptive decision-making systems. It is an innovation for the robot to learn the policy directly from demonstration trajectories to adapt to similar state spaces that may appear in the future. We aim to develop a method for directly learning navigation behavior from demonstration trajectories without defining the environment and attention models, by using the concepts of Generative Adversarial Imitation Learning (GAIL) and Sequence Generative Adversarial Network (SeqGAN). The proposed SeqGAIL model in this thesis allows the robot to reproduce the desired behavior in different situations. In which, an adversarial net is established, and the Feature Counts Errors reduction is utilized as the forcing objective for the Generator. The refinement measure is taken to solve the instability problem. In addition, we proposed to use the Rapidly-exploring Random Tree* (RRT*) with pre-trained weights to generate adequate demonstration trajectories in dynamic environment as the training data, and this idea can effectively overcome the difficulty of acquiring huge training data.

Cette thèse s'intéresse au problème de la navigation autonome au long cours de robots mobiles à roues dans des environnements dynamiques. Elle s'inscrit dans le cadre du projet FUI Air-Cobot. Ce projet, porté par Akka Technologies, a vu collaborer plusieurs entreprises (Akka, Airbus, 2MORROW, Sterela) ainsi que deux laboratoires de recherche, le LAAS et Mines Albi. L'objectif est de développer un robot collaboratif (ou cobot) capable de réaliser l'inspection d'un avion avant le décollage ou en hangar. Différents aspects ont donc été abordés : le contrôle non destructif, la stratégie de navigation, le développement du système robotisé et de son instrumentation, etc. Cette thèse répond au second problème évoqué, celui de la navigation. L'environnement considéré étant aéroportuaire, il est hautement structuré et répond à des normes de déplacement très strictes (zones interdites, etc.). Il peut être encombré d'obstacles statiques (attendus ou non) et dynamiques (véhicules divers, piétons, ...) qu'il conviendra d'éviter pour garantir la sécurité des biens et des personnes. Cette thèse présente deux contributions. La première porte sur la synthèse d'un asservissement visuel permettant au robot de se déplacer sur de longues distances (autour de l'avion ou en hangar) grâce à une carte topologique et au choix de cibles dédiées. De plus, cet asservissement visuel exploite les informations fournies par toutes les caméras embarquées. La seconde contribution porte sur la sécurité et l'évitement d'obstacles. Une loi de commande basée sur les spirales équiangulaires exploite seulement les données sensorielles fournies par les lasers embarqués. Elle est donc purement référencée capteur et permet de contourner tout obstacle, qu'il soit fixe ou mobile. Il s'agit donc d'une solution générale permettant de garantir la non collision. Enfin, des résultats expérimentaux, réalisés au LAAS et sur le site d'Airbus à Blagnac, montrent l'efficacité de la stratégie développée. / This thesis is directed towards the autonomous long range navigation of wheeled robots in dynamic environments. It takes place within the Air-Cobot project. This project aims at designing a collaborative robot (cobot) able to perform the preflight inspection of an aircraft. The considered environment is then highly structured (airport runway and hangars) and may be cluttered with both static and dynamic unknown obstacles (luggage or refueling trucks, pedestrians, etc.). Our navigation framework relies on previous works and is based on the switching between different control laws (go to goal controller, visual servoing, obstacle avoidance) depending on the context. Our contribution is twofold. First of all, we have designed a visual servoing controller able to make the robot move over a long distance thanks to a topological map and to the choice of suitable targets. In addition, multi-camera visual servoing control laws have been built to benefit from the image data provided by the different cameras which are embedded on the Air-Cobot system. The second contribution is related to obstacle avoidance. A control law based on equiangular spirals has been designed to guarantee non collision. This control law, based on equiangular spirals, is fully sensor-based, and allows to avoid static and dynamic obstacles alike. It then provides a general solution to deal efficiently with the collision problem. Experimental results, performed both in LAAS and in Airbus hangars and runways, show the efficiency of the developed techniques.

No description available.

Motion is one of the most important features describing an image sequence. Motion estimation has been widely applied in structure from motion, vision-based navigation and many other fields. However, real-time motion estimation remains a challenge because of its high computational expense. The traditional CPU-based scheme cannot satisfy the power, size and computation requirements in many applications. With the availability of new parallel architectures such as FPGAs and GPUs, applying these new technologies to computer vision tasks such as motion estimation has been an active research field in recent years. In this dissertation, FPGAs have been applied to real-time motion estimation for their outstanding properties in computation power, size, power consumption and reconfigurability. It is believed in this dissertation that simply migrating the software-based algorithms and mapping them to a specific architecture is not enough to achieve good performance. Accuracy is usually compromised as the cost of migration. Improvement and optimization at the algorithm level are critical to performance. To improve motion estimation on the FPGA platform and prove the effectiveness of the method, three main efforts have been made in the dissertation. First, a lightweight tensor-based algorithm has been designed which can be implemented in a fully pipelined structure. Key factors determining the algorithm performance are analyzed from the simulation results. Second, an improved algorithm is then developed based on the analyses of the first algorithm. This algorithm applies a ridge estimator and temporal smoothing in order to improve the accuracy. A structure composed of two pipelines is designed to accommodate the new algorithm while using reasonable hardware resources. Third, a hardware friendly algorithm is developed to analyze the optical flow field and detect obstacles for unmanned ground vehicle applications. The motion component is de-rotated, de-translated and postprocessed to detect obstacles. All these steps can be efficiently implemented in FPGAs. The characteristics of the FPGA architecture are taken into account in all development processes of these three algorithms. This dissertation also discusses some important perspectives for FPGA-based design in different chapters. These perspectives include software simulation and optimization at the algorithm development stage, hardware simulation and test bench design at the hardware development stage. They are important and particular for the development of FPGA-based computer vision algorithms. The experimental results have shown that the proposed motion estimation module can perform in real-time and achieve over 50% improvement in the motion estimation accuracy compared to the previous work in the literature. The results also show that the motion field can be reliably applied to obstacle detection tasks.

Today a group of automated guided vehicles at Toyota Material Handling Manufacturing Sweden detect and avoid objects primarily by using 2D-LiDAR, with shortcomings being the limitation of only scanning the area in a 2D plane and missing objects close to the ground. Several dynamic obstacles exist in the environment of the vehicles. Protruding forks are one such obstacle, impossible to detect and avoid with the current choice of sensor and its placement. This thesis investigates possible solutions and limitations of using a single RGB camera for obstacle detection, tracking, and avoidance. The obstacle detection uses the deep learning model YOLOv5s. A solution for semi-automatic data gathering and labeling is designed, and pre-trained weights are chosen to minimize the amount of labeled data needed. Two different approaches are implemented for the tracking of the object. The YOLOv5s detection is the foundation of the first, where 2D-bounding boxes are used as measurements in an Extended Kalman Filter (EKF). Fiducial markers build up the second approach, used as measurements in another EKF. A state lattice motion planner is designed to find a feasible path around the detected obstacle. The chosen graph search algorithm is ARA*, designed to initially find a suboptimal path and improve it if time allows.  The detection works successfully with an average precision of 0.714. The filter using 2D-bounding boxes can not differentiate between a clockwise and counterclockwise rotation, but the performance is improved when a measurement of rotation is included. Using ARA* in the motion planner, the solution sufficiently avoids the obstacles.

Virtual visualization of mobile robot analytical trajectories while avoiding moving obstacles is presented in this thesis as a very helpful technique to properly display and communicate simulation results. Analytical solutions to the path planning problem of mobile robots in the presence of obstacles and a dynamically changing environment have been presented in the current robotics and controls literature. These techniques have been demonstrated using two-dimensional graphical representation of simulation results. In this thesis, the analytical solution published by Dr. Zhihua Qu in December 2004 is used and simulated using a virtual visualization tool called VRML.

No description available.

No description available.

No description available.

No description available.