This paper proposes a learning-based behavior generation approach for automated vehicles which is adapted sequentially. Instead of engineering behavioral policies for a variety of individual traffic situations by hand, our approach concentrates on a general problem description which is adjusted using a learning algorithm that successively derives safe actions as an outcome. Recent approaches apply Reinforcement Learning techniques for this problem using Markov Decision Processes (MDP). Our approach benefits from a trajectory planning module that uses an optimal control approach and generates realistic trajectories. Further, the trajectory planning module is exploited for the exploration in solving the adaption of the action selection problem. The task of action selection for merging into a roundabout as an exemplary traffic situation is examined. The contributions of this paper are the usage of an underlying optimization-based trajectory generation module and the evaluation of convergence of the adapted behavior, also for real-world data.

Learning Convolutional Neural Networks (CNN) is commonly carried out by plain supervised gradient descent. With sufficient training data, this leads to very competitive results for visual recognition tasks when starting from a random initialization. When the amount of labeled data is limited, CNNs reveal their strong dependence on large amounts of training data. However, recent results have shown that a well chosen optimization starting point can be beneficial for convergence to a good generalizing minimum. This starting point was mostly found using unsupervised feature learning techniques such as sparse coding or transfer learning from related recognition tasks. In this work, we compare these two approaches against a simple patch based initialization scheme and a random initialization of the weights. We show that pre-training helps to train CNNs from few samples and that the correct choice of the initialization scheme can push the network's performance by up to 41% compared to random initialization.

In rural areas, wildlife animal road crossings are a threat to both the driver and the wildlife population. Since most accidents take place at night, recent night vision driver assistance systems are supporting the driver by automatically detecting animals on infrared camera imagery. After detecting an animal on the roadside, the orientation towards the road can give a first cue for an upcoming trajectory prediction. This paper describes an novel classification-based scheme for nighttime animal orientation estimation from single infrared images. Our system classifies already detected animals, in particular deer, as being either oriented left, right or back/front. We propose an approach based on Convolutional Neural Networks which learns multiple stages of invariant features in a supervised end-to-end fashion. Experiments show that our method outperforms baseline methods like HOG/SVM or boosted Haar-stumps on this task.

Non-negative Matrix Factorization is a technique for decomposing large data sets into bases and code words, where all entries of the occurring matrices are non-negative. A recently proposed technique also incorporates sparseness constraints, in such a way that the amount of nonzero entries in both bases and code words becomes controllable. This paper extends the Non-negative Matrix Factorization with Sparseness Constraints. First, a modification of the optimization criteria ensures fast inference of the code words. Thus, the approach is real-time capable for use in time critical applications. Second, in case a teacher signal is associated with the samples, it is considered in order to ensure that inferred code words of different classes can be well distinguished. Thus, the derived bases generate discriminative code words, which is a crucial prerequisite for training powerful classifiers. Experiments on natural image patches show, similar to recent results in the field of sparse coding algorithms, that Gabor-like filters are minimizing the reconstruction error while retaining inference capabilities. However, applying the approach with incorporation of the teacher signal to handwritten digits yields morphologically completely different bases, while achieving superior classification results.