Accelerating Training of Reinforcement Learning-Based Construction Robots in Simulation Using Demonstrations Collected in Virtual Reality
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The application of construction robots is crucial to mitigate challenges faced by the construction industry, such as labor shortages and low productivity. Reinforcement learning (RL) enables robots to take actions based on observed states, improving flexibility over traditional robots pre-programmed to follow determined sequences of instructions. However, RL-based control is time-consuming to train, hindering the wide adoption of RL-based construction robots. This paper proposes an approach that utilizes expert demonstrations collected from virtual reality to accelerate the RL training of construction robots. For evaluation, we implement the approach for the task of window pickup and installation on a virtual construction site. In our experiment, out of 10 RL agents trained using virtual expert demonstrations, 7 agents converge to an optimal policy faster than the baseline RL agent trained without demonstrations by around 40 epochs, which proves adding expert demonstrations can effectively accelerate the training of robots learning construction tasks.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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 it