Kunsoo Huh

Hayeon O, Chanuk Yang, Kunsoo Huh

In autonomous driving, 3D object detection provides more precise information for downstream tasks, including path planning and motion estimation, compared to 2D object detection. In this paper, we propose SeSame: a method aimed at enhancing semantic information in existing LiDAR-only based 3D object detection. This addresses the limitation of existing 3D detectors, which primarily focus on object presence and classification, thus lacking in capturing relationships between elemental units that constitute the data, akin to semantic segmentation. Experiments demonstrate the effectiveness of our method with performance improvements on the KITTI object detection benchmark. Our code is available at https://github.com/HAMA-DL-dev/SeSame

Chanuk Yang, Hayeon O, Kunsoo Huh

This paper proposes a novel algorithm for vehicle speed-aided monocular visual-inertial localization using a topological map. The proposed system aims to address the limitations of existing methods that rely heavily on expensive sensors like GPS and LiDAR by leveraging relatively inexpensive camera-based pose estimation. The topological map is generated offline from LiDAR point clouds and includes depth images, intensity images, and corresponding camera poses. This map is then used for real-time localization through correspondence matching between current camera images and the stored topological images. The system employs an Iterated Error State Kalman Filter (IESKF) for optimized pose estimation, incorporating correspondence among images and vehicle speed measurements to enhance accuracy. Experimental results using both open dataset and our collected data in challenging scenario, such as tunnel, demonstrate the proposed algorithm's superior performance in topological map generation and localization tasks.

Dongchan Kim, Kunsoo Huh

This paper presents a hybrid motion planning strategy that combines a deep generative network with a conventional motion planning method. Existing planning methods such as A* and Hybrid A* are widely used in path planning tasks because of their ability to determine feasible paths even in complex environments; however, they have limitations in terms of efficiency. To overcome these limitations, a path planning algorithm based on a neural network, namely the neural Hybrid A*, is introduced. This paper proposes using a conditional variational autoencoder (CVAE) to guide the search algorithm by exploiting the ability of CVAE to learn information about the planning space given the information of the parking environment. A non-uniform expansion strategy is utilized based on a distribution of feasible trajectories learned in the demonstrations. The proposed method effectively learns the representations of a given state, and shows improvement in terms of algorithm performance.

Hayoung Kim, Dongchan Kim, Gihoon Kim et al.

This paper presents online-capable deep learning model for probabilistic vehicle trajectory prediction. We propose a simple encoder-decoder architecture based on multi-head attention. The proposed model generates the distribution of the predicted trajectories for multiple vehicles in parallel. Our approach to model the interactions can learn to attend to a few influential vehicles in an unsupervised manner, which can improve the interpretability of the network. The experiments using naturalistic trajectories at highway show the clear improvement in terms of positional error on both longitudinal and lateral direction.