Mastering data complexity for autonomous driving with adaptive point clouds for urban environments

Yin, Hang; Berger, Christian

doi:10.1109/ivs.2017.7995901

Cited by 13 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Directly converting 3D point cloud data into a 2D image will inevitably result in the loss of information, therefore some methods have utilized the LiDAR scanner's operating principle and targeted the raw data from the LiDAR sensor, which is known as packet data. Yin et al [15] focused on this raw packet data and on the LiDAR system's data format. In a previous study [3], we proposed converting LiDAR packet data into a 2D matrix and then using an existing image compression method to compress the data.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Streaming Point Cloud Compression for 3D LiDAR Sensor Using U-Net

Takeuchi

Carballo

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Point cloud data from LiDAR sensors is the currently the basis of most L4 autonomous driving systems. Sharing and storing point clouds will also be important for future applications, such as accident investigation or V2V/V2X networks. Due to the huge volume of data involved, storing point clouds collected over long periods of time and transmitting point clouds in real-time are difficult tasks, making compression an indispensable step before storing or transmitting. Previous streaming point cloud compression methods, such as octree compression or video compression-based approaches, have difficulty compressing this data in real-time into very small volumes with low information loss. To reduce temporal redundancy efficiently and rapidly, in this paper we propose a real-time streaming point cloud data compression method using U-net. By utilizing raw packet data from LiDAR sensors, we can store 3D point cloud information losslessly in a 2D matrix, and convert streaming point cloud data into a video-like format. By designating some frames as reference frames and then using U-net to interpolate the remaining LiDAR frames, we can greatly reduce temporal redundancy. Our use of U-net was inspired by a video interpolation approach employed in another study. Noise in LiDAR data is a big issue which significantly affects network training and compression results. In this paper, we propose a padding strategy to alleviate the negative impact of this noise. As a result of these improvements, our proposed method can outperform octree compression, MPEG-based compression and our previously proposed SLAM-based compression method.

show abstract

Section: Related Workmentioning

confidence: 99%

Real-Time Streaming Point Cloud Compression for 3D LiDAR Sensor Using U-Net

Takeuchi

Carballo

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This way, obstacle detection and tracking steps in the perception pipeline can be performed faster. The computational speedup is also dependent on the data structures adapted and used to store the data, we refer the readers to work done in [44]. Updatability.…”

Section: Motivationmentioning

confidence: 99%

Real-Time Dynamic Object Detection for Autonomous Driving Using Prior 3D-Maps

Kiran

Roldão

Irastorza

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Lidar has become an essential sensor for autonomous driving as it provides reliable depth estimation. Lidar is also the primary sensor used in building 3D maps which can be used even in the case of low-cost systems which do not use Lidar. Computation on Lidar point clouds is intensive as it requires processing of millions of points per second. Additionally there are many subsequent tasks such as clustering, detection, tracking and classification which makes real-time execution challenging. In this paper, we discuss real-time dynamic object detection algorithms which leverages previously mapped Lidar point clouds to reduce processing. The prior 3D maps provide a static background model and we formulate dynamic object detection as a background subtraction problem. Computation and modeling challenges in the mapping and online execution pipeline are described. We propose a rejection cascade architecture to subtract road regions and other 3D regions separately. We implemented an initial version of our proposed algorithm and evaluated the accuracy on CARLA simulator.

show abstract

“…Such shapes can only be effectively represented using 3D models [ 39 ], as it is impossible to reflect all the necessary details of such an environment using merely 2D and 2.5D maps, which are used in many terrestrial applications [ 40 ]. Unfortunately, the precise representation of 3D models using point clouds is a highly resource-consuming proposition, especially for real-time applications [ 41 ]. We use a triangular-mesh-based map representation and localize our robot relative to this triangular mesh map because it allows for much more compact representation without significant loss of accuracy [ 42 , 43 ], is more effective for vehicle navigation than other map representations [ 44 ] such as Elevation Mapping [ 45 ], OctoMap [ 46 ] or VoxBlox [ 47 ] and is more convenient for texture mapping [ 48 ].…”

Section: Introductionmentioning

confidence: 99%

Real-Time Parallel-Serial LiDAR-Based Localization Algorithm with Centimeter Accuracy for GPS-Denied Environments

Niedzwiedzki

Niewola

Lipiński

et al. 2020

Sensors

View full text Add to dashboard Cite

In this paper, we introduce a real-time parallel-serial algorithm for autonomous robot positioning for GPS-denied, dark environments, such as caves and mine galleries. To achieve a good complexity-accuracy trade-off, we fuse data from light detection and ranging (LiDAR) and an inertial measurement unit (IMU). The proposed algorithm’s main novelty is that, unlike in most algorithms, we apply an extended Kalman filter (EKF) to each LiDAR scan point and calculate the location relative to a triangular mesh. We also introduce three implementations of the algorithm: serial, parallel, and parallel-serial. The first implementation verifies the correctness of our innovative approach, but is too slow for real-time execution. The second approach implements a well-known parallel data fusion approach, but is still too slow for our application. The third and final implementation of the presented algorithm along with the state-of-the-art GPU data structures achieves real-time performance. According to our experimental findings, our algorithm outperforms the reference Gaussian mixture model (GMM) localization algorithm in terms of accuracy by a factor of two.

show abstract

Mastering data complexity for autonomous driving with adaptive point clouds for urban environments

Cited by 13 publications

References 6 publications

Real-Time Streaming Point Cloud Compression for 3D LiDAR Sensor Using U-Net

Real-Time Streaming Point Cloud Compression for 3D LiDAR Sensor Using U-Net

Real-Time Dynamic Object Detection for Autonomous Driving Using Prior 3D-Maps

Real-Time Parallel-Serial LiDAR-Based Localization Algorithm with Centimeter Accuracy for GPS-Denied Environments

Contact Info

Product

Resources

About