We present an approach for learning spatial traversability maps for driving in complex, urban environments based on an extensive dataset demonstrating the driving behaviour of human experts. The direct end-to-end mapping from raw input data to cost bypasses the effort of manually designing parts of the pipeline, exploits a large number of data samples, and can be framed additionally to refine handcrafted cost maps produced based on manual hand-engineered features. To achieve this, we introduce a maximum-entropy-based, non-linear inverse reinforcement learning (IRL) framework which exploits the capacity of fully convolutional neural networks (FCNs) to represent the cost model underlying driving behaviours. The application of a high-capacity, deep, parametric approach successfully scales to more complex environments and driving behaviours, while at deployment being run-time independent of training dataset size. After benchmarking against state-of-the-art IRL approaches, we focus on demonstrating scalability and performance on an ambitious dataset collected over the course of 1 year including more than 25,000 demonstration trajectories extracted from over 120 km of urban driving. We evaluate the resulting cost representations by showing the advantages over a carefully, manually designed cost map and furthermore demonstrate its robustness towards systematic errors by learning accurate representations even in the presence of calibration perturbations. Importantly, we demonstrate that a manually designed cost map can be refined to more accurately handle corner cases that are scarcely seen in the environment, such as stairs, slopes and underpasses, by further incorporating human priors into the training framework.

This paper presents an end-to-end approach for tracking static and dynamic objects for an autonomous vehicle driving through crowded urban environments. Unlike traditional approaches to tracking, this method is learned end-to-end, and is able to directly predict a full unoccluded occupancy grid map from raw laser input data. Inspired by the recently presented DeepTracking approach ([1], [2]), we employ a recurrent neural network (RNN) to capture the temporal evolution of the state of the environment, and propose to use Spatial Transformer modules to exploit estimates of the egomotion of the vehicle. Our results demonstrate the ability to track a range of objects, including cars, buses, pedestrians, and cyclists through occlusion, from both moving and stationary platforms, using a single learned model. Experimental results demonstrate that the model can also predict the future states of objects from current inputs, with greater accuracy than previous work.

We present the first pipeline for real-time volumetric surface reconstruction and dense 6DoF camera tracking running purely on standard, off-the-shelf mobile phones. Using only the embedded RGB camera, our system allows users to scan objects of varying shape, size, and appearance in seconds, with real-time feedback during the capture process. Unlike existing state of the art methods, which produce only point-based 3D models on the phone, or require cloud-based processing, our hybrid GPU/CPU pipeline is unique in that it creates a connected 3D surface model directly on the device at 25Hz. In each frame, we perform dense 6DoF tracking, which continuously registers the RGB input to the incrementally built 3D model, minimizing a noise aware photoconsistency error metric. This is followed by efficient key-frame selection, and dense per-frame stereo matching. These depth maps are fused volumetrically using a method akin to KinectFusion, producing compelling surface models. For each frame, the implicit surface is extracted for live user feedback and pose estimation. We demonstrate scans of a variety of objects, and compare to a Kinect-based baseline, showing on average ∼ 1.5cm error. We qualitatively compare to a state of the art point-based mobile phone method, demonstrating an order of magnitude faster scanning times, and fully connected surface models.

This paper explores the idea of reducing a robot's energy consumption while following a trajectory by turning off the main localisation subsystem and switching to a lowerpowered, less accurate odometry source at appropriate times. This applies to scenarios where the robot is permitted to deviate from the original trajectory, which allows for energy savings. Sensor scheduling is formulated as a probabilistic belief planning problem. Two algorithms are presented which generate feasible perception schedules: the first is based upon a simple heuristic; the second leverages dynamic programming to obtain optimal plans. Both simulations and real-world experiments on a planetary rover prototype demonstrate over 50% savings in perception-related energy, which translates into a 12% reduction in total energy consumption. I. INTRODUCTIONRobots require energy to operate. Yet they only have access to limited energy storage during missions. As we extend the reach of autonomous systems to operate in remote locations, over long distances and for long periods of time, energy considerations are becoming increasingly important. To date, these considerations are often brought to bear in schemes where trajectories or speed profiles are optimised to minimise the energy required for actuation (see, for example, [1], [2], [3]). Here we take a different, yet complementary, approach in considering the energy expenditure for sensing (and, implicitly, computation) associated with navigation. In particular, our goal is to activate the perception system only as required to maintain the vehicle within a given margin around a predetermined path. As the main navigation sensors are switched off and the robot reverts to a lower-powered, less accurate odometry source for parts of the trajectory, any associated computation will also be reduced, leading to further savings in energy.Naively, such perception schedules could be constructed by switching sensors on and off randomly or according to, for example, a fixed frequency. This does, however, suffer the drawback that no heed is paid to drift in the robot's position with respect to the original trajectory: it may not be desirable to deviate by more than an allowed margin from the predetermined route. This arises, for example, in a planetary exploration scenario when conducting long traverses over featureless terrain. Other possible considerations include traversability, obstacles, and the robustness of the localisation system to deviations from the original path. Such naive approaches would also need to be tuned to individual trajectories as savings would depend significantly on trajectory shape. In this work we present two approaches which explicitly account for drift and trajectory shape (though the