This paper describes the motivation, design and implementation of a Perception Validation System (PVS), a system for measuring the outdoor perception performance of an autonomous vehicle. The PVS relies on using large amounts of real world data and ground truth information to quantify performance aspects such as the rate of false positive or false negative detections of an obstacle detection system. Our system relies on a relational database infrastructure to achieve a high degree of flexibility in the type of analyses it can support. We discuss the main steps required for going from raw data to numerical estimates describing the performance of the perception system, including the generation of ground truth information and the safe speed metric we found to be most useful for comparing the perception system's outputs to the ground truth data. We present results illustrating some of the analyses that can be completed using the Perception Validation System.

Tangible augmented reality (AR) systems imbue physical objects with the ability to act and respond in new ways. In particular, physical objects and gestures made with them gain meaning that does not exist outside the tangible AR environment. The existence of this new set of possible actions and outcomes is not always apparent, making it necessary to learn new movements or gestures. Addressing this opportunity, we present visual hints, which are graphical representations in AR of potential actions and their consequences in the augmented physical world.Visualhintsenable discovery, learning, and completion of gestures and manipulation in tangible AR. Here, we discuss our investigation of a variety of representations of visual hints and methods for activating them. We then describe a specific implementation that supports gestures developed for a tangible AR user interface to an electronic field guide for botanists, and present results from a pilot study.