"Cognitive Computing" has initiated a new era in computer science. Cognitive computers are not rigidly programmed computers anymore, but they learn from their interactions with humans, from the environment and from information. They are thus able to perform amazing tasks on their own, such as driving a car in dense traffic, piloting an aircraft in difficult conditions, taking complex financial investment decisions, analysing medical-imaging data, and assist medical doctors in diagnosis and therapy. Cognitive computing is based on artificial intelligence, image processing, pattern recognition, robotics, adaptive software, networks and other modern computer science areas, but also includes sensors and actuators to interact with the physical world. Cognitive computers – also called "intelligent machines" – are emulating the human cognitive, mental and intellectual capabilities. They aim to do for human mental power (the ability to use our brain in understanding and influencing our physical and information environment) what the steam engine and combustion motor did for muscle power. We can expect a massive impact of cognitive computing on life and work. Many modern complex infrastructures, such as the electricity distribution grid, railway networks, the road traffic structure, information analysis (big data), the health care system, and many more will rely on intelligent decisions taken by cognitive computers. A drawback of cognitive computers will be a shift in employment opportunities: A raising number of tasks will be taken over by intelligent machines, thus erasing entire job categories (such as cashiers, mail clerks, call and customer assistance centres, taxi and bus drivers, pilots, grid operators, air traffic controllers, …). A possibly dangerous risk of cognitive computing is the threat by “super intelligent machines” to mankind. As soon as they are sufficiently intelligent, deeply networked and have access to the physical world they may endanger many areas of human supremacy, even possibly eliminate humans. Cognitive computing technology is based on new software architectures – the “cognitive computing architectures”. Cognitive architectures enable the development of systems that exhibit intelligent behaviour.

International crash statistics indicate a decrease of bicycle crashes, but at a slower pace compared to total crash numbers. The share of crashes with involved cyclists is above the modal share (see [1] for an overview). Depending on sources, types of analyses, and geographic regions, crash statistics suggest high rates of singlebike crashes and crashes between cyclists and other vulnerable road users (VRUs) [2], while cars are opponents in more than half of all fatal crashes in the European Union [3]. The design of th.e road environment is of particular relevance for crash risks. A study from London found three times higher injury odds for cyclists at intersections [4]. Connected and automated vehicles (CAV) are frequently said to increase the safety level in road traffic since they are less prone to human errors [5]. This might hold true in transport systems with little complexity, such as highways [6]. However, when it comes to complex situations in multimodal systems with multiple interactions between different road users, such as intersections in urban environments, existing solutions are not sufficient yet in terms of protecting VRUs. ... In order to contribute to the safety of VRUs in the interplay with CAVs in current systems, we propose a geospatial model, which delineates potential interaction risk zones from high definition (HD) maps and enriching these zones with additional information. These enriched risk zones are then provided as standardized OGC web service, which can be integrated in V2X systems. With this, we contribute to the visibility, and thus the safety of VRUs in connected transport systems. From a methodological point of view, the proposed model is a first step in integrating spatial context and semantic information explicitly into V2X communication. [From: Introduction]

The localization of autonomous ground vehicles in dense urban environments poses a challenge. Applications in classical outdoor robotics rely on the availability of GPS systems in order to estimate the position. However, the presence of complex building structures in dense urban environments hampers a reliable localization based on GPS. Alternative approaches have to be applied In order to tackle this problem. This thesis proposes an approach which combines observations of a single perspective camera and odometry in a probabilistic framework. In particular, the localization in the space of appearance is addressed. First, a topological map of reference places in the environment is built. Each reference place is associated with a set of visual features. A feature selection is carried out in order to obtain distinctive reference places. The topological map is extended to a hybrid representation by the use of metric information from Geographic Information Systems (GIS) and satellite images. The localization is solved in terms of the recognition of reference places. A particle lter implementation incorporating this and the vehicle's odometry is presented. The proposed system is evaluated based on multiple experiments in exemplary urban environments characterized by high building structures and a multitude of dynamic objects.

Cognitive Computing' has initiated a new era in computer science. Cognitive computers are not rigidly programmed computers anymore, but they learn from their interactions with humans, from the environment and from information. They are thus able to perform amazing tasks on their own, such as driving a car in dense traffic, piloting an aircraft in difficult conditions, taking complex financial investment decisions, analysing medical-imaging data, and assist medical doctors in diagnosis and therapy. Cognitive computing is based on artificial intelligence, image processing, pattern recognition, robotics, adaptive software, networks and other modern computer science areas, but also includes sensors and actuators to interact with the physical world. Cognitive computers – also called 'intelligent machines' – are emulating the human cognitive, mental and intellectual capabilities. They aim to do for human mental power (the ability to use our brain in understanding and influencing our physical and information environment) what the steam engine and combustion motor did for muscle power. We can expect a massive impact of cognitive computing on life and work. Many modern complex infrastructures, such as the electricity distribution grid, railway networks, the road traffic structure, information analysis (big data), the health care system, and many more will rely on intelligent decisions taken by cognitive computers. A drawback of cognitive computers will be a shift in employment opportunities: A raising number of tasks will be taken over by intelligent machines, thus erasing entire job categories (such as cashiers, mail clerks, call and customer assistance centres, taxi and bus drivers, pilots, grid operators, air traffic controllers, …). A possibly dangerous risk of cognitive computing is the threat by “super intelligent machines” to mankind. As soon as they are sufficiently intelligent, deeply networked and have access to the physical world they may endanger many areas of human supremacy, even possibly eliminate humans. Cognitive computing technology is based on new software architectures – the “cognitive computing architectures”. Cognitive architectures enable the development of systems that exhibit intelligent behaviour.:Introduction 5 1. Applying the Subsumption Architecture to the Genesis Story Understanding System – A Notion and Nexus of Cognition Hypotheses (Felix Mai) 9 2. Benefits and Drawbacks of Hardware Architectures Developed Specifically for Cognitive Computing (Philipp Schröppe)l 19 3. Language Workbench Technology For Cognitive Systems (Tobias Nett) 29 4. Networked Brain-based Architectures for more Efficient Learning (Tyler Butler) 41 5. Developing Better Pharmaceuticals – Using the Virtual Physiological Human (Ben Blau) 51 6. Management of existential Risks of Applications leveraged through Cognitive Computing (Robert Richter) 61

The localization of autonomous ground vehicles in dense urban environments poses a challenge. Applications in classical outdoor robotics rely on the availability of GPS systems in order to estimate the position. However, the presence of complex building structures in dense urban environments hampers a reliable localization based on GPS. Alternative approaches have to be applied In order to tackle this problem. This thesis proposes an approach which combines observations of a single perspective camera and odometry in a probabilistic framework. In particular, the localization in the space of appearance is addressed. First, a topological map of reference places in the environment is built. Each reference place is associated with a set of visual features. A feature selection is carried out in order to obtain distinctive reference places. The topological map is extended to a hybrid representation by the use of metric information from Geographic Information Systems (GIS) and satellite images. The localization is solved in terms of the recognition of reference places. A particle lter implementation incorporating this and the vehicle's odometry is presented. The proposed system is evaluated based on multiple experiments in exemplary urban environments characterized by high building structures and a multitude of dynamic objects.

The paper proposes the management of bus disruption (e.g. fleet failure) and maintain a resilient transportation system through a synergy between shared autonomous vehicles and the existing public transport system based on the organizational structure and demand characteristics. The methodology is applied to the region of Rennes (France) and its surroundings.