Development of Autonomous Drones for Adaptive Obstacle Avoidance in Real World Environments (original) (raw)
Related papers
Towards Obstacle Avoidance and Autonomous UAV Operation
Proc. of AHS …, 2011
This paper presents the status and progress of the ongoing work directed towards the development and implementation of autonomous navigation algorithms for Micro Aerial Vehicles (MAV). The method proposed is founded on a mapping methodology, which is supported by a laser scan matching algorithm and virtual occupancy grid method. Navigation and path planning is performed by means of an extended and optimized version of the potential field approach. This paper contains a description of the methodology along with initial results from both simulations and experiments that demonstrate the ability to navigate around corners from start to goal positions as well as mapping realistic corridor environments using a ground platform. A notable advantage of the current methodology is the separation between the MAV's model and the navigation algorithm, which makes them suitable for various rotary wing (and other) platforms.
Evolving Philosophies on Autonomous Obstacle/Collision Avoidance of Unmanned Aerial Vehicles
2011
Much of the benefits of deploying unmanned aerial vehicles can be derived from autonomous missions. For such missions, however, sense-and-avoid capability (i.e., the ability to detect potential collisions and avoid them) is a critical requirement. Collision avoidance can be broadly classified into global and local path-planning algorithms, both of which need to be addressed in a successful mission. Whereas global path planning (which is mainly done offline) broadly lays out a path that reaches the goal point, local collision-avoidance algorithms, which are usually fast, reactive, and carried out online, ensure safety of the vehicle from unexpected and unforeseen obstacles/collisions. Even though many techniques for both global and local collision avoidance have been proposed in the recent literature, there is a great interest around the globe to solve this important problem comprehensively and efficiently and such techniques are still evolving. This paper presents a brief overview of a few promising and evolving ideas on collision avoidance for unmanned aerial vehicles, with a preferential bias toward local collision avoidance.
A Control of Multiple Drones for Automatic Collision Avoidance
Information Technology Journal, 2015
Unmanned Aerial Vehicles (UAV) are often used in hazardous missions. Many models, sizes and types are available for different usages. A well-known type of UAV is “drone”, which is controlled by a remote controller via radio wave. These drones have internal memory and use battery for energy source. At present, commercial off-the-shelf drones are subject to human manual control, which can only control one drone at a time. In order to enable autonomous control of multiple drones, the detection of objects around the drones and collision avoidance mechanism are necessary. This paper presents a control of multiple drones for automatic collision avoidance by using a detection device that is composed of ultrasonic sensors and an embedded control device to detect objects in four directions. A collision avoidance algorithm is designed based on the object detection to enable automatic avoidance. The result has shown that the detection device and the collision avoidance algorithm can work with ...
Mini Quadcopter on Obstacle Avoidance and Mapping for Indoor Tasks in Narrow Areas
2024
The ability of an autonomous vehicle, found in an unknown environment while blind, to navigate its way and avoid obstacles by scanning the surroundings is a crucial focus of researchers. The term "obstacle avoidance" in this context encompasses the detection of objects, decision-making based on detection information, and execution of movements. This literature review examines the obstacle avoidance capabilities of autonomous aerial vehicles designed to operate in challenging environments such as narrow, inadequate lightening, and dusty conditions, particularly for urban search and rescue operations. Comparisons are made among sensors for object detection and algorithms for motion decision-making. In the study, the system modeling phase involved the utilization of ready-made obstacle avoidance blocks provided by Simulink, coupled with the design of a drone controller to integrate VFH outputs as inputs for control. The performance of the system was visualized in three dimensions and in real-time, with comparisons drawn between paths followed with and without obstacle avoidance features. Optimization efforts were directed towards minimizing time, with recorded durations facilitating comparative analyses. Key findings included the identification of optimal parameters such as the cost function, lidar range, and proportional gains.
Obstacle Avoidance System for UAVs using Computer Vision
AIAA Infotech @ Aerospace, 2015
The purpose of this research is to develop an obstacle avoidance system for use on small, fixed-wing Uninhabited Aerial Vehicles (UAVs). In order to detect and avoid obstacles, computer based vision algorithms will be implemented with an automatic flight control system. Images of obstacles are captured using forward facing, externally mounted cameras. Obstacles will include moving and non-moving objects within the flight path of the UAV, which will be detected through the use of optical flow and feature-tracking methods. 1. Motivation and Goals for Research UAVs have the potential to replace inhabited aircraft for many civilian and military applications, which include, but not limited to, disaster relief assistance, search and rescue, and combat zone intelligence gathering. They have lower operating costs and pose minimal risk to human pilots. However,
International journal of computer applications, 2018
The need of Unmanned Aerial Systems (UASs) is expanding day by day as it can be used in both public and military environments. As the need for UAV is growing, there is an expansion inthe requirement for more reliable, authentic, efficient, optimized and strong vehicles that are capable for executing various operations. The need of such systems is mainly by the Militaries that continue to desire more UAV functionalities for diverse operations and tasks that can be performed all over the world. To have a continuous advancement in the field of autonomous UAV control system many cogent research works has been performed. A large amount of work is focused on the subsets of UAS control such as path planning algorithms, control of small UAV and autonomy. As various markets are amplifying, the necessity to have such systems with capability to adapt according to introduced tasks, sensing elements, and surroundings will drive requirements. They can be used in several ways in various models, sizes and types according to the needs of various operations. The most common Unmanned Aerial Vehicle is "drone"that can be operated by remote controllers using radio waves. These UAVs normally contains internal memory and uses battery power as a means to an energy source. In the current scenario, commercial drones are normally subjected to manual control by human that can control only one drone at a point of time. In accordance to set up a controlling system of multiple drones, collision avoidance mechanism and the detection of objects around these UAV systems is very necessary. This paper discusses about the UAVs requirements and capabilities along with path planning algorithms. It also provides some problems associated with a UAV system along with its improvement areas.
Modeling and Prototyping a Modular, Low-Cost Collision Avoidance System for UAVs
2021 IEEE Aerospace Conference (50100), 2021
Many challenges arise when attempting to use unmanned aerial vehicles (UAVs) in indoor environments, such as the lack of a GPS signal for use in navigation and the smaller margin of error in movements. Typically, those challenges are addressed by using a collision avoidance system. However, most commercially available collision avoidance systems are expensive, limited in suppliers, and are restricted to use on a specific platform. Additionally, some of the collision avoidance systems choose to forego obstacle detection in one or more directions, usually the upward direction. This work proposes that it is possible to develop a custom, low-cost collision avoidance system with modular capabilities, allowing it to be adapted to any UAV platform. The feasibility of the proposed system was determined by creating a single-direction prototype that was implemented on a small quadcopter and tested by flying the quadcopter towards a wall at slow speeds. To develop the system's control algorithm a model of a quadcopter was built. Two different control algorithms were developed and tested via simulation with the model, and the better performing algorithm was implemented in the prototype. The feasibility of the proposed collision avoidance system is promising with the prototype able to prevent the quadcopter from colliding with a wall. However, further refinement in the methodology and techniques used to develop the system is needed to improve performance and reliability of the system, especially as obstacle detection is added in other directions of motion.
IJERT-Vision based Obstacle Avoidance System for Autonomous Aerial Systems
International Journal of Engineering Research and Technology (IJERT), 2021
https://www.ijert.org/vision-based-obstacle-avoidance-system-for-autonomous-aerial-systems https://www.ijert.org/research/vision-based-obstacle-avoidance-system-for-autonomous-aerial-systems-IJERTCONV9IS03154.pdf Here we present this topic to explore the idea of how autonomous unmanned aerial vehicles are creating significant transformational revolution with the collision avoidance algorithms. And, also, how we could strategically create more advanced ways to improve and improvise the functioning of sensors used in aerial robotics by replacing them with the cost-efficient camera modules and apply AI or even non-AI enabled computer vision. The primary aim of the project is to look for various systems which could help autonomous drones in detecting obstacles with the help of computer vision, and accordingly letting the UAV to maneuver in order to prevent collision.
Robot Vision: Obstacle-Avoidance Techniques for Unmanned Aerial Vehicles
IEEE Robotics & Automation Magazine, 2000
I n this article, a vision-based technique for obstacle avoidance and target identification is combined with haptic feedback to develop a new teleoperated navigation system for underactuated aerial vehicles in unknown environments. A three-dimensional (3-D) map of the surrounding environment is built by matching the keypoints among several images, which are acquired by an onboard camera and stored in a buffer together with the corresponding estimated odometry. Hence, based on the 3-D map, a visual identification algorithm is employed to localize both obstacles and the desired target to build a virtual field accordingly. A bilateral control system has been developed such that an operator can safely navigate in an unknown environment and perceive it by means of a vision-based haptic force-feedback device. Experimental tests in an indoor environment verify the effectiveness of the proposed teleoperated control. Vision-Based Obstacle-Avoidance Techniques Interest in unmanned aerial vehicles (UAVs) has increased due to the wide range of their application fields, which include surveillance, rescue, and inspection. So far, UAVs have mainly been used outdoors with the support of a global positioning system (GPS) for navigation purposes. However, when UAVs are flying indoors in an unknown and unstructured environment, GPS information will not be available. Therefore, in such situations, different navigation and obstacle-avoidance techniques have been investigated using onboard sensors such as lasers, sonars, cameras, radars, and inertial measurement units (IMUs) that give a perception of the environment. Obstacle avoidance is a core issue since any autonomous navigation system must preserve the safety of both the UAV and the surrounding environment. Several approaches can be found to address this problem. In [1] and [2], radar-based navigation and obstacle avoidance are implemented, while a laser range finder for obstacle detection is employed in [3]. The main drawbacks are the high power consumption and weight of these sensors.