Under a fully observable 40 × 40 occupancy-grid model, dynamic obstacles are represented as time-varying occupied cells that are refreshed before each decision. The agent then acts in a four-neighbour discrete action space. This paper proposes a lightweight residual deep Q-network (ResNet-DQN) for fast trajectory planning in this setting. Two-dimensional normalized coordinates and a six-channel compact map representation are used as state inputs. The training protocol combines planner-labelled behavioural pretraining and decaying training-only guidance with Double DQN targets, validity-gated action selection, reward shaping, Huber loss, gradient clipping and hybrid soft-hard target-network updates. Under a matched five-seed protocol using seeds 11, 37, 71, 101 and 102, ResNet-DQN achieved a mean success rate of 94.8% and a complete decision latency of 0.912 ms on an NVIDIA GeForce RTX 4090 D. The present model is limited to four-direction discrete motion and does not represent continuous robot kinematics or predict obstacle trajectories.
Fast ResNet-DQN trajectory planning algorithm for dynamic grid environments. // Iranian Journal of Science and Technology, Transactions of Civil Engineering. – 2026. – DOI: 10.1007/s42452-026-09903-4. – URL: https://link.springer.com/article/10.1007/s42452-026-09903-4
(2026). Fast ResNet-DQN trajectory planning algorithm for dynamic grid environments. Iranian Journal of Science and Technology, Transactions of Civil Engineering. https://doi.org/10.1007/s42452-026-09903-4
“Fast ResNet-DQN trajectory planning algorithm for dynamic grid environments.” Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2026. doi:10.1007/s42452-026-09903-4.
@article{article35488,
title = {Fast ResNet-DQN trajectory planning algorithm for dynamic grid environments},
journal = {Iranian Journal of Science and Technology, Transactions of Civil Engineering},
year = {2026},
doi = {10.1007/s42452-026-09903-4},
url = {https://link.springer.com/article/10.1007/s42452-026-09903-4}
}
Duckweed is an important resource for bioenergy and animal feed, yet conventional cultivation methods often suffer from uneven nutrient distribution and excess water consumption. This study introduced novel superhydrophobic (SHP)...
Rivers are essential for sustaining human activities, ecological balance, and water supply. However, urbanization remains a major factor controlling quality of rivers. Several studies have been conducted to examine hydrological...
Natural language interaction and physical environment understanding are two core challenges for improving the autonomy of home service robots. This paper provides a systematic review of the evolution from traditional robotic methods to...
To address the difficulty of achieving dynamic coordination between particulate removal efficiency and energy consumption under wide load fluctuations, this study develops digital twins (DTs) system for an electrostatic precipitator...
Bioinspired composite design moves structural material development from a chemistry first to a structure first approach, in which hierarchical architecture delivers combinations of stiffness, toughness, impact resistance, and durability...