OPAL: Outlier-Preserved Microscaling Quantization A ccelerator for Generative Large Language Models

To overcome the burden on the memory size and bandwidth due to ever-increasing size of large language models (LLMs), aggressive weight quantization has been recently studied, while lacking research on quantizing activations. In this paper, we present a hardware-software co-design method that results in an energy-efficient LLM accelerator, named OPAL, for generation tasks. First of all, a novel activation quantization method that leverages the microscaling data format while preserving several outliers per sub-tensor block (e.g., four out of 128 elements) is proposed. Second, on top of preserving outliers, mixed precision is utilized that sets 5-bit for inputs to sensitive layers in the decoder block of an LLM, while keeping inputs to less sensitive layers to 3-bit. Finally, we present the OPAL hardware architecture that consists of FP units for handling outliers and vectorized INT multipliers for dominant non-outlier related operations. In addition, OPAL uses log2-based approximation on softmax operations that only requires shift and subtraction to maximize power efficiency. As a result, we are able to improve the energy efficiency by 1.6~2.2x, and reduce the area by 2.4~3.1x with negligible accuracy loss, i.e., <1 perplexity increase.

Quantitative 3D Map Accuracy Evaluation Hardware and Algorithm for LiDAR(-Inertial) SLAM

Accuracy evaluation of a 3D pointcloud map is crucial for the development of autonomous driving systems. In this work, we propose a user-independent software/hardware system that can quantitatively evaluate the accuracy of a 3D pointcloud map acquired from LiDAR(-Inertial) SLAM. We introduce a LiDAR target that functions robustly in the outdoor environment, while remaining observable by LiDAR. We also propose a software algorithm that automatically extracts representative points and calculates the accuracy of the 3D pointcloud map by leveraging GPS position data. This methodology overcomes the limitations of the manual selection method, that its result varies between users. Furthermore, two different error metrics, relative and absolute errors, are introduced to analyze the accuracy from different perspectives. Our implementations are available at: https://github.com/SangwooJung98/3D_Map_Evaluation

Co-RaL: Complementary Radar-Leg Odometry with 4-DoF Optimization and Rolling Contact

Robust and accurate localization in challenging environments is becoming crucial for SLAM. In this paper, we propose a unique sensor configuration for precise and robust odometry by integrating chip radar and a legged robot. Specifically, we introduce a tightly coupled radar-leg odometry algorithm for complementary drift correction. Adopting the 4-DoF optimization and decoupled RANSAC to mmWave chip radar significantly enhances radar odometry beyond the existing method, especially z-directional even when using a single radar. For the leg odometry, we employ rolling contact modeling-aided forward kinematics, accommodating scenarios with the potential possibility of contact drift and radar failure. We evaluate our method by comparing it with other chip radar odometry algorithms using real-world datasets with diverse environments while the datasets will be released for the robotics community. https://github.com/SangwooJung98/Co-RaL-Dataset

Imaging radar and LiDAR image translation for 3-DOF extrinsic calibration

The integration of sensor data is crucial in the field of robotics to take full advantage of the various sensors employed. One critical aspect of this integration is determining the extrinsic calibration parameters, such as the relative transformation, between each sensor. The use of data fusion between complementary sensors, such as radar and LiDAR, can provide significant benefits, particularly in harsh environments where accurate depth data is required. However, noise included in radar sensor data can make the estimation of extrinsic calibration challenging. To address this issue, we present a novel framework for the extrinsic calibration of radar and LiDAR sensors, utilizing CycleGAN as amethod of image-to-image translation. Our proposed method employs translating radar bird-eye-view images into LiDAR-style images to estimate the 3-DOF extrinsic parameters. The use of image registration techniques, as well as deskewing based on sensor odometry and B-spline interpolation, is employed to address the rolling shutter effect commonly present in spinning sensors. Our method demonstrates a notable improvement in extrinsic calibration compared to filter-based methods using the MulRan dataset.

TRansPose: Large-Scale Multispectral Dataset for Transparent Object

Transparent objects are encountered frequently in our daily lives, yet recognizing them poses challenges for conventional vision sensors due to their unique material properties, not being well perceived from RGB or depth cameras. Overcoming this limitation, thermal infrared cameras have emerged as a solution, offering improved visibility and shape information for transparent objects. In this paper, we present TRansPose, the first large-scale multispectral dataset that combines stereo RGB-D, thermal infrared (TIR) images, and object poses to promote transparent object research. The dataset includes 99 transparent objects, encompassing 43 household items, 27 recyclable trashes, 29 chemical laboratory equivalents, and 12 non-transparent objects. It comprises a vast collection of 333,819 images and 4,000,056 annotations, providing instance-level segmentation masks, ground-truth poses, and completed depth information. The data was acquired using a FLIR A65 thermal infrared (TIR) camera, two Intel RealSense L515 RGB-D cameras, and a Franka Emika Panda robot manipulator. Spanning 87 sequences, TRansPose covers various challenging real-life scenarios, including objects filled with water, diverse lighting conditions, heavy clutter, non-transparent or translucent containers, objects in plastic bags, and multi-stacked objects. TRansPose dataset can be accessed from the following link: https://sites.google.com/view/transpose-dataset

Asynchronous Multiple LiDAR-Inertial Odometry using Point-wise Inter-LiDAR Uncertainty Propagation

In recent years, multiple Light Detection and Ranging (LiDAR) systems have grown in popularity due to their enhanced accuracy and stability from the increased field of view (FOV). However, integrating multiple LiDARs can be challenging, attributable to temporal and spatial discrepancies. Common practice is to transform points among sensors while requiring strict time synchronization or approximating transformation among sensor frames. Unlike existing methods, we elaborate the inter-sensor transformation using continuous-time (CT) inertial measurement unit (IMU) modeling and derive associated ambiguity as a point-wise uncertainty. This uncertainty, modeled by combining the state covariance with the acquisition time and point range, allows us to alleviate the strict time synchronization and to overcome FOV difference. The proposed method has been validated on both public and our datasets and is compatible with various LiDAR manufacturers and scanning patterns. We open-source the code for public access at https://github.com/minwoo0611/MA-LIO.