Automated Roadside COW Animal Detection and Collision Prevention System

Automated Roadside COW Animal Detection and Collision Prevention System

Unsettled Topics in Obstacle Detection for Autonomous Agricultural Vehicles

Agricultural vehicles often drive along the same terrain day after day or year after year. Yet, they still must detect if a moveable object, such as another vehicle or an animal, happens to be on their path or if environmental conditions have caused muddy spots or washouts. Obstacle detection is one of the major missing pieces that can remove humans from highly automated agricultural machines today and enable the autonomous vehicles of the future. Unsettled Topics in Obstacle Detection for Autonomous Agricultural Vehicles examines the challenges of environmental object detection and collision prevention, including air obscurants, holes and soft spots, prior maps, vehicle geometry, standards, and close contact with large objects.

MODELLING OF VTS SUPERVISOR BY ALGORITHM BASED ON PETRI NET: CASE STUDY OF DOVER INCIDENT

The paper deals with collision prevention problem in maritime transport in the area of the narrow canals with predefined routes. The Dover incident, which is analysed and described in the paper, has shown that the control of the passage of ships through the critical areas must be upgraded with an automatic supervising system, which warns the human operator of incorrect ship motion and help the operator to make the right and timely decision. The general idea is to improve the safety of navigation by introduction of automatic collision prevention based on automated supervisor helping to human operator in Vessel Traffic System (VTS) control centre. The VTS supervisor automatically monitors marine traffic by using data from Automatic Radar Plotting Aid (ARPA) radar and others sensors. Such supervisor detects real time and Course Over Ground (COG) of the vessel entering a particular sector, and then estimates the required time for vessel’s passage into another sector. VTS supervisor compares the real time and estimated time of passage of the specific ship through particular sector as a part of surveillance area. In addition, it compares and monitors the deviation of the course during transition of zones (sectors). If significant difference for both values are occurred VTS supervisor triggers a time alarm or a course alarm respectively. In the paper authors have modelled and simulated collision prevention with performed by the alarm actions of VTS supervisor improved with algorithm module based on hybrid Petri net formalism and Visual Object Net ++ tool.

A Study on an Enhanced Autonomous Driving Simulation Model Based on Reinforcement Learning Using a Collision Prevention Model

This paper set out to revise and improve existing autonomous driving models using reinforcement learning, thus proposing a reinforced autonomous driving prediction model. The paper conducted training for a reinforcement learning model using DQN, a reinforcement learning algorithm. The main aim of this paper was to reduce the time spent on training and improve self-driving performance. Rewards for reinforcement learning agents were developed to mimic human driving behavior as much as possible. High rewards were given for greater distance travelled within lanes and higher speed. Negative rewards were given when a vehicle crossed into other lanes or had a collision. Performance evaluation was carried out in urban environments without pedestrians. The performance test results show that the model with the collision prevention model exhibited faster performance improvement within the same time compared to when the model was not applied. However, vulnerabilities to factors such as pedestrians and vehicles approaching from the side were not addressed, and the lack of stability in the definition of compensation functions and limitations with respect to the excessive use of memory were shown.

MUSSOL: A Micro-Uas to Survey Ship Cargo hOLds

Because of their high maneuverability and fast deployment times, aerial robots have recently gained popularity for automating inspection tasks. In this paper, we address the visual inspection of vessel cargo holds, aiming at safer, cost-efficient and more intensive visual inspections of ships by means of a multirotor-type platform. To this end, the vehicle is equipped with a sensor suite able to supply the surveyor with imagery from relevant areas, while the control software is supporting the operator during flight with enhanced functionalities and reliable autonomy. All this has been accomplished in the context of the supervised autonomy (SA) paradigm, by means of extensive use of behaviour-based high-level control (including obstacle detection and collision prevention), all specifically devised for visual inspection. The full system has been evaluated both in laboratory and in real environments, on-board two different vessels. Results show the vehicle effective for the referred application, in particular due to the inspection-oriented capabilities it has been fitted with.

Investigations on digitalization for sustainable machine tools and forming technologies

Progress in applied research for sustainable machine tools and forming technologies bases upon industrial and environmental requirements for resource efficiency. Relevant technical trends base upon impact studies and applied research projects on the lifecycle resource consumption for manufacturing processes and systems. This paper gives an overview about a unified methodological approach of the evaluation of resource efficiency of machine tools. It answers the scientific question on sustainability: which technological parameters and machine tool characteristics lead to their lowest resource consumption/losses and part manufacturing costs. Therefore, the method allows to consider them as an energy-information model, in which the transformation of any forms and types of energy, material, and information takes place. It is shown that innovative hollow shaft forming technologies become sustainable alternatives to cutting technologies. A smart factory uses digitalization, manufacturing data management, and self-learning methods for resource efficiency. Sustainable production requires robust and error-free machining processes. Therefore, a collision prevention system protects machining centers and work pieces from collisions in real time will be presented. The gathered information about the product and its properties as well as manufacturing data builds a digital twin and enables a prediction of the resource consumption in smart factories.