One of the fundamental challenges in the design of perception systems for autonomous vehicles is validating the performance of each algorithm under a comprehensive variety of operating conditions. In the case of vision-based semantic segmentation, there are known issues when encountering new scenarios that are sufficiently different to the training data. In addition, even small variations in environmental conditions, such as illumination and precipitation, can affect the classification performance of the segmentation model. Given the reliance on visual information, these effects often translate into poor semantic pixel classification which can potentially lead to catastrophic consequences when driving autonomously. This paper presents a novel method for analyzing the robustness of semantic segmentation models and provides a number of metrics to evaluate the classification performance over a variety of environmental conditions. The process incorporates an additional sensor (lidar) to automate the process and improve the system integrity, eliminating the need for labor-intensive hand labeling of validation data. The experimental results are presented based on multiple datasets collected at different times of the year with different environmental conditions. We extract the ”Road” class using the lidar to demonstrate the concepts, but this could be extended to other classes with different feature detection algorithms. These results show that the semantic segmentation performance varies depending on the weather, camera parameters, and existence of shadows. The results also demonstrate how the metrics can be used to compare and validate the performance after making improvements to a model, and compare the performance of different networks.

Vision systems are widely used in autonomous vehicle systems due to the rich information that camera sensors provide of the surrounding environment. This paper presents an automatic algorithm to obtain the drivable path of a vehicle operating in urban roads with or without clear lane markings. The developed system projects trajectories obtained during human operation of the vehicle and utilizes these to generate automatic labels for training a semantic based path prediction model. The system segments an urban scenario into 13 categories including vehicles, pedestrian, undrivable road, other categories relevant to urban roads, and a new class for a path proposal. The drivable path information is essential particularly in unstructured scenarios, and is critical for an intelligent vehicle system to make sound driving decisions. The path proposal category is a car-width drivable lane estimated to be safe to drive for the vehicle under consideration. The data collection, model training and inference process requires only images from a monocular camera and odometry from a low-cost IMU combined with a wheel encoder. The algorithm has been successfully demonstrated on the Sydney University campus, which is a challenging environment without clear road markings. The algorithm was demonstrated to run in real-time, proving its applicability for intelligent vehicles.

To navigate through urban roads, an automated vehicle must be able to perceive and recognize objects in a three-dimensional environment. A high level contextual understanding of the surroundings is necessary to execute accurate driving maneuvers. This paper presents a novel approach to build three dimensional semantic octree maps from lidar scans and the output of a convolutional neural network (CNN) to obtain the labels of the environment. We present a heuristic method to associate uncertainties to the labels from the images based on a combination of the labels themselves, score maps retrieved by the CNN and the raw images. These uncertainties and the camera-lidar calibration parameters for multiple cameras are considered in the projection of the labels and their uncertainties into the point cloud. Every labeled lidar scan works as an input to an octree map building algorithm that calculates and updates the label probabilities of the voxels in the map. This paper also presents a qualitative and quantitative evaluation of accuracy, analyzing projection in single lidar scans and complete maps built with our probabilistic octree framework.