Testing autonomous vehicles in real-world scenarios would pose a threat to people and property alike. A safe alternative is to simulate these scenarios and test to ensure that the resulting programs can work in real-world scenarios. Moreover, in order to detect a moral dilemma situation quickly, the vehicle should be able to identify objects in real-time while driving. Toward this end, this thesis investigates the use of cross-platform training for neural networks that perform visual identification of common objects in driving scenarios. Here, the object detection algorithm Faster R-CNN is used. The hypothesis is that it is possible to train a neural network model to detect objects from two different domains, simulated or physical, using transfer learning. As a proof of concept, an object detection model is trained on image datasets extracted from CARLA, a virtual driving environment, via transfer learning. After bringing the total loss factor to 0.4, the model is evaluated with an IoU metric. It is determined that the model has a precision of 100% and 75% for vehicles and traffic lights respectively. The recall is found to be 84.62% and 75% for the same. It is also shown that this model can detect the same classes of objects from other virtual environments and real-world images. Further modifications to the algorithm that may be required to improve performance are discussed as future work.
The future of driving is largely headed towards autonomous vehicles, and this is clear with companies such as Tesla, Waymo, and even tech giant Apple. Many professionals predict that autonomous vehicles will likely be commercially available and legal to use in some places by the late 2020s [15]. There are some benefits to the rapid development of autonomous vehicle controllers, such as more independence for those who can’t drive due to impairments, the potential for reduced traffic, as well as possibly decreasing the number of accidents. Though these are promising prospects, there are ethical concerns regarding the implementation of such technology. The goal of this thesis is to provide an introductory literature review that discusses the history of autonomous vehicles, different levels of autonomy, ethical considerations in autonomous systems, and prior work on characterizing human driving behaviors and implementing these behaviors with autonomous vehicle controllers. Finally, recommendations are proposed for data collection on human driving behaviors in an ongoing NSF-funded project at Arizona State University, “Embodiment of Human Values Profiles in Autonomous Vehicles via Psychomimetic Controller Design.”
Cancer therapy selects for cancer cells resistant to treatment, a process that is fundamentally evolutionary. To what extent, however, is the evolutionary perspective employed in research on therapeutic resistance and relapse? We analyzed 6,228 papers on therapeutic resistance and/or relapse in cancers and found that the use of evolution terms in abstracts has remained at about 1% since the 1980s. However, detailed coding of 22 recent papers revealed a higher proportion of papers using evolutionary methods or evolutionary theory, although this number is still less than 10%. Despite the fact that relapse and therapeutic resistance is essentially an evolutionary process, it appears that this framework has not permeated research. This represents an unrealized opportunity for advances in research on therapeutic resistance.