Landing Zone Identification Using A Hardware-accelerated Deep Learning Module
Bibliographic record
Abstract
This work develops a deep learning-based autonomous Landing Zone (LZ) identification module for a Vertical Take- Off and Landing (VTOL) drone using colored Light Detection and Ranging (LiDAR) point cloud data. "ConvPoint", a top-performing neural network (NN) architecture of the Semantic3D.net pointcloud segmentation benchmark leaderboard, was chosen as the reference architecture for the development. A classification method based on the terrain geometry characteristics is used for automatic labeling of the datasets followed by manual adjustment of label through visual observation. The automatic labelling method selected is a state-of-the-art LZ detection method reported in literature which also serves as the baseline for comparative evaluation. Point clouds captured by the Intelligent Systems Laboratory (ISL), Memorial University of Newfoundland (MUN) and online point cloud datasets were used to perform network training and comparative evaluation of the methods. The results signify the enhanced capability of deep learning based methods on handling both geometry and color information for LZ estimation, and the ability to perform LZ estimations making use of hardware accelerator modules. The deep learning base methods were capable of achieving accuracies up to 94% for datasets that contain water bodies where the classical approach had poor predictive capability due to the reliance on only geometric information. The proposed LZ detection algorithm was run on a reconfigurable hardware-accelerated module to evaluate the real-time feasibility of the approach which currently is capable of 10238 points per second processing speed on Jetson AGX Xavier dedicated hardware.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".