Autonomous navigation

PROJECT ARROW
TECHNICAL CONCEPT MEMORANDUM
NAVIGATION SYSTEM (TRL1 - 2)

Relativistic Astrodynamics:
A Real-Time N-Body Navigation & Telemetry Framework

RANGIS-U — Relativistic Autonomous Navigational &
Gravitational Integration System (Universal)

This paper defines the architecture for the Relativistic Autonomous Navigational & Gravitational Integration System (RANGIS). Designed for near-light-speed (relativistic) spacecraft, this system solves the N-body problem in real-time by integrating non-Euclidean gravitational perturbations from massive sources (stellar, planetary, neutron stars, and black holes) while dynamically updating the spacecraft's rest mass and relativistic momentum.

Students with Santa Clara University (SCU) and the Monterey Bay Aquarium Research Institute (MBARI) are developing an innovative platform for shallow water bathymetry. Bathymetry data is used to analyze the geography, ecosystem, and health of marine habitats. Current methods for shallow water measurements typically involve large manned vessels that are costly to operate and that may pose a danger to themselves and the environment in shallow, semi-navigable waters. Small vessels, however, are prone to disturbances by shallow water waves, tides, and currents, thereby requiring more instrumentation and computation to accurately process bathymetric data. The SCU/MBARI autonomous surface vessel, SeaWASP, is designed to operate safely and stably in waters as shallow as 1 m without significant manned support in order to produce cost-effective and high-quality bathymetric maps.The SeaWASP design introduces several key design innovations in order to provide high-quality maps with a platform ...

Autonomous vehicles stand at the forefront of technological advancement, promising to alter fundamentally how we navigate, with implications for everything from inter city logistics to daily commuting. The expected benefits of this technology include advancements in safety protocols, streamlining of traffic flow, and superior comfort for those onboard. However, the path to widespread integration is paved with signifi cant barriers that need to be overcome. This article provides a thorough review of the historical development of autonomous driving technologies, beginning with the early vehicles powered by internal combustion engines and extending to the present excite ment over Apple’s Project Titan. Rumors suggest this particular project involves the creation of an electric vehicle fully capable of operating without traditional manual controls, including steering wheels and seats. The article concludes by spotlighting the potential for autonomous vehicles to enable disruptive changes in the transport sector, while also considering the myriad of challenges that must be surmounted for their adoption on a grand scale.

Existing controllers for sailboat robots are usually developed for speed performances and for long straight lines. In this context, the accuracy is not the main concern. In this paper, we consider the tight slalom problem which requires accuracy. We propose a feedback-linearization based method combined with a vector field approach to control the sailboat. Some simulations show that the robot is able to perform the slalom without missing any gate.

The Cool Robot is a four-wheel-drive, solarpowered autonomous vehicle designed to support summertime science campaigns in Antarctica and Greenland. We deployed the robot at Summit Camp, Greenland, during 2005 to validate its power budget and to assess its unique control system that matches solar power input with power demand as the robot drives over rough terrain. The 61-kg robot drove continuously at 0.78 m/s on soft snow, its 160-W average power demand met by solar power alone under clear skies when sun elevation exceeded 16°. The power-control system reliably matched input with demand as insolation changed during the tests. A simple GPS waypoint-following algorithm provided reliable autonomous navigation over periods of 5 -8 hours. The data validate our design models and indicate that the Cool Robot will exceed its design goal of carrying a 15-kg payload 500 km in two weeks on the Antarctic plateau.

This paper presents a hybrid path planning algorithm for the design of autonomous vehicles such as mobile robots. The hybrid planner is based on Potential Field method and Voronoi Diagram approach and is represented with the ability of concurrent navigation and map building. The system controller (Look-ahead Control) with the Potential Field method guarantees the robot generate a smooth and safe path to an expected position. The Voronoi Diagram approach is adopted for the purpose of helping the mobile robot to avoid being trapped by concave environment while exploring a route to a target. This approach allows the mobile robot to accomplish an autonomous navigation task with only an essential exploration between a start and goal position. Based on the existing topological map the mobile robot is able to construct sub-goals between predefined start and goal, and follows a smooth and safe trajectory in a flexible manner when stationary and moving obstacles co-exist.

Este trabalho explora o uso da Inteligência Artificial e Reconhecimento de Padrões para compor o processo de estimativa da altura de voo de um vetor aéreo (drone, avião, helicóptero, entre outros) em um processo de navegação autônoma. Verifica-se neste trabalho que, desde que os parâmetros da câmera sejam conhecidos, a rede neural convolucional empregada seja satisfatoriamente treinada e o reconhecimento de padrões de alvos de interesse com suas métricas sejam determinados, a estimativa da altura de voo resultante pode ser empregada como método de navegação autônoma nos vetores aéreos e, assim, compor mais uma alternativa para mitigar a negação ou interferência nos sinais GNSS. A rede neural convolucional treinada identificou 90% dos alvos de interesse e a estimativa da altura de voo apresentou resultados satisfatórios.

Intelligence, smart sensors are very much required to link with naval systems in current era to take autonomous decision making and enhancing the capabilities of existing systems. Artificial Intelligence (AI) based algorithms and deepcatch models are making predominant changes in the existing sensing technologies with much more precision in a real-time environment which will revolutionize Navy. But there are lot of challenges and features during the incorporation of AI in the performance upgradation of naval platforms. The latest improvement in the AI proactive real-time intelligence algorithms exceeding the human intelligence capability with the help of Merlin Deep Learning (DL) methods/models to extract and analyse the sensor parameters reading from the naval systems. Nowadays, Internet of Things (IoT) sensors installed with Reinforcement Learning (RL) algorithms are used to detect and process the object information in the naval systems. This paper discuss the key factors involved in the incorporation and upgradation of current sensor systems with their issues, challenges, benefits and features. The ideas and concepts presented in this paper are merely intended to provoke and stimulate thinking about human-machine interaction teams that would function on the AI-enabled battlefield of the future.

The work is dedicated to a comprehensive comparative analysis of autonomous navigation methods for swarms of unmanned aerial vehicles. It examines the conceptual differences, technical advantages, and critical disadvantages of centralized and decentralized control approaches in complex environments. Based on aggregated data, key algorithmic foundations (Swarm Intelligence, APF), issues of scalability, fault tolerance, the impact of communication interference, and cybersecurity threats are analyzed. Prospects for the use of adaptive hybrid architectures and Edge Computing to enhance the overall efficiency and autonomy of the swarm are presented. The results of the analysis can be used for the informed selection of swarm control architecture when designing systems for civilian and military applications.

This paper introduces a novel modular vehicle architecture based on a distributed array of identical wheel units designed to achieve robust mobility across heterogeneous environments, including terrestrial and aquatic domains. In contrast to conventional vehicle systems that employ environment-specific propulsion mechanisms (e.g., tires, propellers, or tracks), the proposed architecture establishes a unified locomotion framework in which propulsion, load distribution, and stability are achieved through a large population of geometrically and functionally identical rotary elements.
The vehicle is configured as an articulated multi-segment structure, wherein each segment integrates a high-density arrangement of wheel units, enabling scalable force generation and distributed contact mechanics. Analytical modeling demonstrates that the total traction force scales linearly with the number of active wheel units (Ftotal=N⋅Funit), while the effective ground pressure decreases inversely with the total contact area (P=W/(N⋅Aunit), indicating a substantial reduction in terrain stress compared to conventional low-wheel-count configurations. For representative configurations with N=40–100 units, the model predicts ground pressure reductions exceeding 60–85% relative to standard four-wheel systems under equivalent loading conditions.
Furthermore, the distributed architecture inherently provides fault tolerance, as system functionality is preserved under partial actuator failure, with performance degradation proportional to the fraction of failed units rather than catastrophic loss. Hydrodynamic analysis suggests that rotating wheel units can generate thrust in aquatic environments through paddle-like interaction, offering a simplified alternative to dedicated marine propulsion, albeit with reduced efficiency compared to optimized propellers.
The study also examines the limitations of extending wheel-based actuation to aerial mobility, concluding that additional aerodynamic structures are required to achieve sufficient lift. Overall, the proposed system represents a scalable, redundant, and multi-functional mobility paradigm, with potential applications in multi-terrain transport, amphibious systems, and resilient robotic platforms. The presented framework provides a theoretical foundation for future experimental validation and optimization of distributed locomotion systems.

Obstacle detection and avoidance systems as a research area have been gaining attention lately, but the issues of blind areas and ground/small obstacle problems have received little interest, especially indoor autonomous navigation. In this work, we present an environment-sweep technique for obstacle detection and a fuzzy algorithm for ground obstacle avoidance, surface elevation estimation, and obstacle classification for indoor robots that eliminates the blind area problem. We employed an ultrasonic sensor on a servo for environmental sweep to eliminate blind areas. Results show that the issue of interference was mitigated by the servo sweep technique we propose. Fuzzy algorithms combined with a single ultrasonic sensor reduced the complexity of the robot, thereby improving the robot's performance in real-time. Fuzzy algorithms combined with a differential wheel driving mechanism speed up the inference and actuation process of the proposed system. This is due to fewer rules and a less complex yet effective algorithm design that converts steering angles to different wheel speeds, making the response time much less than the stipulated threshold of 2s as has been determined in previous literature.

Fuzzy set, since its advent has played an important role in control systems and many other area of applications. One of such area is the control of autonomous vehicle. There seem to be some difficulty however, for a new timer trying to get a clear picture of the autonomous navigation problem. To this end, this survey presents a panoramic view of the Intelligent Transportation Systems with some few example of the Advance Driver Assistance Systems and a good discussion on the autonomous systems with its eminent problems. More attention was focused on the fuzzy controllers designed for collision avoidance; as its performance has largely simplified and smoothens the collision avoidance process of an autonomous vehicular system.

Vehicle classification has evolved into a significant subject of study due to its importance in autonomous navigation, traffic analysis, surveillance and security systems, and transportation management. While numerous approaches have been introduced for this purpose, no specific study has been conducted to provide a robust and complete video-based vehicle classification system based on the rear-side view where the camera's field of view is directly behind the vehicle. In this paper we present a stochastic multi-class vehicle classification system which classifies a vehicle (given its direct rear-side view) into one of four classes Sedan, Pickup truck, SUV/Minivan, and unknown. A feature set of tail light and vehicle dimensions is extracted which feeds a feature selection algorithm to define a low-dimensional feature vector. The feature vector is then processed by a Hybrid Dynamic Bayesian Network (HDBN) to classify each vehicle. Results are shown on a database of 169 videos for four classes.

In this article we present the state of the art in the field of autonomous surface ship navigation using machine learning. We discuss the main challenges towards the development of fully autonomous navigation systems with the International Regulations for Preventing Collisions at Sea (COLREGs). Finally, we propose two alternative approaches that are based on machine learning. Existing COLREGs-based navigation and collision avoidance algorithms are based on traditional search-based planning and optimization algorithms. We consider that these approaches are suitable when the problem space is defined completely and rigorously. However, experts believe that is not the case for COLREGs since it leaves many aspects open to the interpretation of the captain. For example, COLREGs expects that any collision avoidance action shall be taken with due regard to the observance of good seamanship, a concept not defined in the convention. Furthermore, many rules are defined using undefined concepts like safe distance, or keywords like early, or substantial, without giving any definition. COLREGs even allow for the rules to be broken to avoid an accident. Due to this, traditional planning approaches may not be able to handle complex scenarios that are underspecified according to COLREGs. An alternative is the use of machine learning (ML), reinforcement learning (RL) and imitation learning (IL) at the core of autonomous navigation systems. Machine learning is known to succeed and outperform traditional approaches specially in vaguely defined problem domains, where it is difficult, if not impossible, to create a full formal specification of the phenomenon under study. We consider this to be the case for COLREGs-based navigation and we conjecture that a ML-based navigation approach can outperform existing search-based and optimization algorithms.

This paper explores the state of the art on to methods to verify and validate navigation algorithms for autonomous surface ships. We perform a systematic mapping study to find research works published in the last 10 years proposing new algorithms for autonomous navigation and collision avoidance and we have extracted what verification and validation approaches have been applied on these algorithms. We observe that most research works use simulations to validate their algorithms. However, these simulations often involve just a few scenarios designed manually. This raises the question if the algorithms have been validated properly. To remedy this, we propose the use of a systematic scenario-based testing approach to validate navigation algorithms extensively.

This paper presents a time-varying Gabor filter bank predictor for use with vehicle tracking via surveillance video. A frame-based 2D Gabor-filter bank is selected as a primary detector for any changes in a given video frame sequence. Detected changes are localized in each frame by fitting a bounding box on the silhouette of the vehicle in the region of interest (ROI). Arbitrary motion of each vehicle is fed to a nonlinear directional predictor in the time axis for estimating the location of the tracked vehicle in the next frame of the video sequence. Real-time traffic-video experimentation dictates that the cone Gabor filter structure is able to tune itself into a selected target and trace it accordingly. This property is highly desirable in the fast and accurate moving vehicle or target tracking purposes in range and intensity driven sensing.

This paper presents a time-varying Gabor filter bank predictor for use with vehicle tracking via surveillance video. A frame-based 2D Gabor-filter bank is selected as a primary detector for any changes in a given video frame sequence. Detected changes are localized in each frame by fitting a bounding box on the silhouette of the vehicle in the region of interest (ROI). Arbitrary motion of each vehicle is fed to a nonlinear directional predictor in the time axis for estimating the location of the tracked vehicle in the next frame of the video sequence. Real-time traffic-video experimentation dictates that the cone Gabor filter structure is able to tune itself into a selected target and trace it accordingly. This property is highly desirable in the fast and accurate moving vehicle or target tracking purposes in range and intensity driven sensing.

As NASA's Artemis program inaugurates a new era of human deep-space exploration, the development of autonomous navigation systems capable of operating independently of Earth-based infrastructure has become an operational imperative. This study investigated the design, implementation, and validation of an integrated artificial intelligence (AI) framework combining convolutional neural network (CNN)–based optical navigation with reinforcement learning (RL)–based trajectory optimization for autonomous guidance, navigation, and control (GNC) in cislunar space, informed by NASA's Artemis II crewed lunar flyby mission launched on April 1, 2026. The study employed a simulation-based experimental design incorporating Circular Restricted Three-Body Problem (CRTBP) dynamics and realistic optical imaging models. A CNN crater detection module processed synthetic navigation imagery against the Robbins (2019) global lunar crater catalog, an Extended Kalman Filter (EKF) fused optical measurements with inertial data for continuous state estimation, and a Proximal Policy Optimization (PPO) reinforcement learning agent computed fuel-optimal trajectory correction maneuvers in real time. Performance was evaluated across 13 quantitative criteria using 200 Monte Carlo simulation runs. The integrated system met 9 of 13 performance criteria, achieving CNN crater detection precision of 0.890, mean average precision of 0.862 at IoU 0.50, crater matching accuracy of 0.866, and position determination within approximately 5 km using purely autonomous onboard processing. The RL controller achieved mean fuel savings of 14.82% over the classical proportional-derivative baseline while simultaneously reducing position tracking error by 22.8%, demonstrating a Pareto improvement in the fuel-accuracy trade space. The system operated within real-time computational constraints (CNN: 0.74 s, EKF: 41 ms, RL: 6.8 ms per cycle), confirming onboard deployment feasibility. The framework is assessed at Technology Readiness Level 3–4 and provides a validated pathway toward AI-enhanced autonomous navigation for Artemis missions and future deep-space exploration in GPS-denied environments.

Motion safety for robotic systems operating in the real world is critical (especially when their size and dynamics make them potentially harmful for themselves or their environment). Motion safety is a taken-for-granted and ill-defined notion in the Robotics literature and the primary contribution of this paper is to propose three safety criteria that helps in understanding a number of key aspects related to the motion safety issue. A number of navigation schemes used by robotic systems operating in the real-world are then evaluated with respect to these safety criteria. It is established that, in all cases, they violate one or several of them. Accordingly, motion safety, especially in the presence of moving objects, cannot be guaranteed (in the sense that these robotic systems may end up in a situation where a collision inevitably occurs later in the future). Finally, it is shown that the concept of Inevitable Collision States introduced in [1] does respect the three above-mentioned safety criteria and therefore offers a theoretical answer to the motion safety issue.

This paper presents a novel navigation architecture for automated carlike vehicles in urban environments. Motion safety is a critical issue in such environments given that they are partially known and highly dynamic with moving objects (other vehicles, pedestrians. . . ). The main feature of the navigation architecture proposed is its ability to make safe motion decisions in real-time, thus taking into account the harsh constraints imposed by the type of environments considered. The architecture is based upon an efficient publish/subscribe middleware system that allows modularity in design and the easy integration of the key functional components required for autonomous navigation, namely perception, localisation, mapping, real-time motion planning and motion tracking. After an overall presentation of the architecture and its main modules, the paper focuses on the "motion" components of the architecture. Experimental results carried out on a simulated Cycab vehicle are presented.

This paper presents a novel navigation architecture for automated carlike vehicles in urban environments. Motion safety is a critical issue in such environments given that they are partially known and highly dynamic with moving objects (other vehicles, pedestrians. . . ). The main feature of the navigation architecture proposed is its ability to make safe motion decisions in real-time, thus taking into account the harsh constraints imposed by the type of environments considered. The architecture is based upon an efficient publish/subscribe-based middleware system that allows modularity in design and the easy integration of the key functional components required for autonomous navigation: perception, localisation, mapping, real-time motion planning and motion tracking. After an overall presentation of the architecture and its main modules, the paper focuses on the "motion" components of the architecture. Experimental results carried out on both a simulation platform and a Cycab vehicle are presented.

Comparé à la rédaction d'une Habilitation à Diriger des Recherches (HDR), la rédaction d'une thèse de doctorat est une simple formalité (pour ainsi dire). Au cours de notre activité de recherche, nous sommes tous amenés à lire un certain nombre de thèses de doctorat. Dans la forme, elles sont relativement proches, et quand vient le temps de rédiger sa propre thèse de doctorat, cette forme standard constitue un guide précieux. Il en est tout autre chose pour l'HDR. Pour commencer, on n'est que rarement amené à en lire dans le cours normal de notre activité de recherche. Peu ou prou, on commence à s'intéresser à ce genre littéraire le jour où l'on décide de rédiger sa propre HDR. On demande alors aux collègues et amis qui sont passés par là de bien vouloir nous montrer leurs oeuvres. On est surpris de constater alors l'étonnante diversité du genre ! Dans sa version "lourde", la HDR s'apparente (pour ceux qui s'en souviennent) à l'ancienne thèse d'état. Dans sa version "légère", c'est une simple collection d'articles. entre ces deux extrêmes, tout est possible. Dans ces conditions, le rédacteur d'une HDR est quelque peu perplexe : que faire ? quel contenu ? quel volume ? quel type de lecteur ? Autant de questions auquel le rédacteur doit apporter soi-même ses propres réponses. Comme d'autres avant nous, c'est ce que nous avons du faire. Nous avons donc fait des choix. Ce n'est certainement pas à l'Histoire de dire si nos choix ont été pertinents, mais plus modestement à nos lecteurs (et surtout à nos rapporteurs). Nous espérons que ce sera le cas. . .

Vehicle navigation in dynamic environments is an important challenge, especially when the motion of the objects populating the environment is unknown. Traditional motion planning approaches are too slow to be applied in realtime to this domain, hence, new techniques are needed. Recently, iterative planning has emerged as a promising approach. Nevertheless, existing iterative methods do not provide a way to estimate the future behaviour of moving obstacles and use the resulting estimates in trajectory computation. This paper presents an iterative planning approach that addresses these two issues. It consists of two complementary methods: 1) a motion prediction method which learns typical behaviours of objects in a given environment. 2) an iterative motion planning technique based on the concept of Velocity Obstacles.

This paper presents a novel navigation architecture for automated carlike vehicles in urban environments. Motion safety is a critical issue in such environments given that they are partially known and highly dynamic with moving objects (other vehicles, pedestrians. . . ). The main feature of the navigation architecture proposed is its ability to make safe motion decisions in real-time, thus taking into account the harsh constraints imposed by the type of environments considered. The architecture is based upon an efficient publish/subscribe middleware system that allows modularity in design and the easy integration of the key functional components required for autonomous navigation, namely perception, localisation, mapping, real-time motion planning and motion tracking. After an overall presentation of the architecture and its main modules, the paper focuses on the "motion" components of the architecture. Experimental results carried out on a simulated Cycab vehicle are presented.

Vehicle navigation in dynamic environments is an important challenge, especially when the motion of the objects populating the environment is unknown. Traditional motion planning approaches are too slow to be applied in real-time to this domain, hence, new techniques are needed. Recently, iterative planning has emerged as a promising approach. Nevertheless, existing iterative methods do not provide a way to estimate the future behavior of moving obstacles and use the resulting estimates in trajectory computation. This paper presents an iterative planning approach that addresses these two issues. It consists of two complementary methods: 1) a motion prediction method which learns typical behaviors of objects in a given environment. 2) an iterative motion planning technique based on the concept of Velocity Obstacles.

Motion safety for robotic systems operating in the real world is critical (especially when their size and dynamics make them potentially harmful for themselves or their environment). Motion safety is a taken-for-granted and ill-defined notion in the Robotics literature and the primary contribution of this paper is to propose three safety criteria that helps in understanding a number of key aspects related to the motion safety issue. A number of navigation schemes used by robotic systems operating in the real-world are then evaluated with respect to these safety criteria. It is established that, in all cases, they violate one or several of them. Accordingly, motion safety, especially in the presence of moving objects, cannot be guaranteed (in the sense that these robotic systems may end up in a situation where a collision inevitably occurs later in the future). Finally, it is shown that the concept of Inevitable Collision States introduced in [1] does respect the three above-mentioned safety criteria and therefore offers a theoretical answer to the motion safety issue.

This paper addresses the problem of navigation of complex systems in dynamic environments under uncertainty. Such environments impose a hard real time constraint. However, computing a complete motion to the goal within a limited time is impossible to achieve in most real situations. The Partial Motion Planning (PMP) approach presented in is used to address the planning problem. However, for real applications, it is important to take into account uncertainty. In this paper, we present an extension to the PMP that accounts for uncertainty in order to plan trajectories robust to the robot's errors. We show that PMP framework is highly suitable to account for these constraints and present original simulation results of robust trajectories for a car-like robot evolving among moving obstacles.

Autonomous navigation in open and dynamic environments is an important challenge, requiring to solve several difficult research problems located on the cutting edge of the state of the art. Bassically, these problems can be classified into three main categories: SLAM in dynamic environments; Detection, characterization, and behavior prediction of the potential moving obstacles; On-line motion planning and safe navigation decision based on world state predictions. This paper addresses some aspects of these problems and presents our latest approaches and results. The solutions we have implemented are mainly based on the followings paradigms: Characterization and motion prediction of the observed moving entities using bayesian programming; Online goal-oriented navigation decisions using the Partial Motion Planning ( ¢¡ £ ) paradigm. we can deal with a single "antecedent cell" when evaluating the occupancy state of a given cell. (i) Choosing the relevant variables and decomposition.

Since the dawn of maritime history, lighthouses and lighted buoys have stood as the steadfast sentinels of the coast—our "coastal fortresses" that serve as the invisible flags of a nation's sovereignty and the psychological anchors of the mariner. For centuries, the IALA information system has provided watch officers with more than mere coordinates; it has offered a legacy of trust, a sense of home, and an awareness of danger passed down from the "ancestors of the sea". However, as the industry navigates a foundational epoch toward Maritime Autonomous Surface Ships (MASS), the extinguishing or absence of these lights creates a perilous gap that traditional methods can no longer bridge alone.
This project plan represents a technological revolution designed to sustain that ancestral spirit within a modern, unified strategic framework. At its core, the system replaces technological clutter with Spatial Intelligence and organized AR/VR interfaces, ensuring that the "Digital Captain"—a synergistic human-AI partnership—perceives the sea not as a maze of data, but with the same clarity and intuition as the captains of old. By introducing Digital DNA (d-DNA), we establish a multidimensional identity recognition system that weaves together human competency and autonomous operational integrity.
The architecture proposed herein is a masterpiece of redundancy. It integrates the "invisible backbone" of submarine cabling, repurposed as active sensor hubs via Distributed Acoustic Sensing (DAS), with high-throughput LEO satellite networks and modernized AIS/LRIT systems. To safeguard this ecosystem, we implement a Cybersecurity Mesh Architecture (CSMA) and Zero Trust protocols, ensuring that every element backs up the other. This is not merely an automation plan; it is the drafting of a future where the guiding light of the past shines through the digital twins of tomorrow, preserving the soul of the sea while conquering the complexities of the silicon age.

L’approche bionique, consistant a donner des proprietes des etres vivants a un robot, est un moyen de diminuer efficacement l’ecart existant entre le fonctionnement d’un operateur humain et d’une machine. Dans cette etude, le comportement d’anticipation visuo-motrice observe chez l’homme lors du controle des deplacements a ete implemente sur une base mobile teleoperee. Cette implementation a ete effectuee suivant deux modes de couplage : orientation de camera asservie a la direction du deplacement (modele “ plate-forme ”) ou bien, direction du deplacement asservie a l’orientation de la camera (modele “ camera ”). Les resultats montrent 1 / qu’un teleoperateur est plus performant lorsqu’il pilote un vehicule pourvu d’anticipation et 2 / que cet avantage est d’autant plus marque suivant le modele “ camera ”. Cela confirme que la cooperation homme-machine beneficie de l’implantation sur l’engin teleopere de modes de couplages visuo-moteurs proches de ceux de l’operateur.

Recently we have presented a system for panoramic depth imaging with a single standard camera. One of the problems of such a system is the fact that we cannot generate a stereo pair of images in real time. This paper presents a possible solution to this problem. Based on a new sensor setup simulations were performed to establish the quality of new results in comparison to results obtained with the old sensor setup. The goal of the paper is to reveal whether the new setup can be used for real time capturing of panoramic depth images and consequently for autonomous navigation of a mobile robot in a room.

In this paper we present a novel approach to legged humanoid navigation on indoor environments using classical probabilistic SLAM methods based on odometry information and laser measurements. We use two small lasers installed in the robot feet to capture the laser data. Odometry is obtained by calculating the position of each feet at every time step. The SLAM problem is solved by using a multi-laser SLAM solution together with a holonomic motion model. Navigation skills also include a path planning module with obstacle avoidance for autonomous navigation in indoor environments, and the whole process is performed without external computational power. Optionally, localization robustness is increased by adding the detection of landmarks using a camera. Results obtained are presented for the 1.5 m tall Reem-B humanoid robot.

For path planning of robots, it is very much essential that the traversability map of regions on which robot is intended to navigate must be available, the robot must have an idea about surface roughness along with obstacles of surface which it traverses. There are various methods for estimation of traversability on rough terrain, but those which are based on deep learning are found to be most efficient. In this work we have followed deep learning approach in which terrain heightmaps are converted into images and the quadruped robot is simulated on various terrains, the part of terrain where robot gets stuck the corresponding image patch is labeled as non-traversable and the part of terrain where robot travels easily the corresponding heightmap image patch is labeled as traversable. Further the data obtain from simulations is feed to CNN classifier which is then used for estimating traversability of unknown terrains for quadruped. In this paper the algorithm is proposed, which uses this trained CNN classifier for making robots traverse through smoother and safer paths.

In this thesis, a simulation package for the Six Degrees of Freedom (6DOF) motion of an underwater vehicle is developed. Mathematical modeling of an underwater vehicle is done and the parameters needed to write such a simulation package are obtained from an existing underwater vehicle available in the literature. Basic equations of motion are developed to simulate the motion of the underwater vehicle and the parameters needed for the hydrodynamic modeling of the vehicle is obtained from the available literature. 6DOF simulation package prepared for the underwater vehicle was developed using the MATLAB environment. S-function hierarchy is developed using the same platform with C++ programming language. With the usage of S-functions the problems related to the speed of the platform have been eliminated. The use of Sfunction hierarchy brought out the opportunity of running the simulation package on other independent platforms and get results for the simulation.M.S. - Master of Scienc

This article presents a review of researches done on the subject of automated tractor. An autonomous tractor is a vehicle that can operate without or with minimal human control, self-propelled and guided automatically along a desired path. The benefits from such a system are useful for agriculture industry by reducing labor cost and time, as well as improving output efficiency by eliminating human errors. Many researches and inventions have been made, with the results ranging from successful, encouraging to some that are impractical for commercial implementation for certain reasons. These implements include sensor, global navigation satellite system, machine vision, laser triangulation, ultrasonic transmitter and geomagnetic controller, as well as actuator and servo motor.

An autonomous robotic platform should be able to perform long-range and long-endurance missions, which energy limitation is one of the most important challenges. Studies show that motion is not the only power consumer. Management of all power resources is therefore important for these systems. Moreover, many applications for control of autonomous platform are being developed and one important aspect is the excess of information, frequently redundant, that imposes a great computational cost in data processing and consequently power consummation. In our previous works, we have proposed a visual-perception system based on an automatic image discarding method as a simple solution to improve the performance of a real-time navigation system. In this paper, we propose a new environment observer method based on Pearson's Correlation Coefficient. This monocular-vision system permits that some logical components may be shut down to save processor energy consumption, and/or to make the CPU...

The task of robot vision is to recognize the geometry of the robot workspace from a digital image. It is our aim to find the relation between the coordinates of a point in the two-dimensional (2D) image and the coordinates of the point in the real threedimensional (3D) robot environment.

This paper presents a new on-line planner for dynamic environments that is based on the concept of Velocity Obstacles (VO). It addresses the issue of motion safety, i.e. avoiding states of inevitable collision, by selecting a proper time horizon for the velocity obstacle. The proper choice of the time horizon ensures that the boundary of the velocity obstacle coincides with the boundary of the set of inevitable collision states. This time horizon is determined by the minimum time it would take the robot to avoid collision, either by stopping or by passing the respective obstacle. The planner generates a near-time optimal trajectory to the goal by selecting at each time step the velocity that minimizes the time-to-go and is out of the velocity obstacle. The planner takes into account the shape, velocity, and path curvature of the obstacle's trajectory. It is demonstrated for on-line motion planning in very crowded static and dynamic environments.