Recent works (e.g., [1,2]) have applied end-to-end DRL to mobile robots, but they often fail when task objectives change (e.g., from “go to point A” to “inspect three zones”). Conversely, classical SLAM + planning pipelines are brittle under perceptual aliasing.
https://github.com/autonomousrobots2026/modular_drl_scheduler References [1] K. Zhu, T. Zhang, “Deep RL for mobile robots in cluttered environments,” Autonomous Robots , vol. 46, pp. 345–360, 2022. [2] J. Schulman et al., “Proximal policy optimization,” arXiv:1707.06347 , 2017. [3] M. Quigley et al., “ROS: an open-source robot operating system,” ICRA workshop, 2009. [4] S. Thrun et al., Probabilistic Robotics , MIT Press, 2005. [5] L. Chen, “Graph-based task allocation for multi-robot systems,” IEEE T-RO , vol. 39, no. 2, pp. 891–907, 2023. LetPub notation: This paper is a simulated example for illustrative purposes. No actual submission to Autonomous Robots has occurred. For real author guidelines, see https://www.springer.com/journal/10514. autonomous robots letpub
Autonomous Navigation and Task Allocation in Unstructured Environments: A Modular Deep Reinforcement Learning Approach Recent works (e
L. Chen¹, M. Kowalski², S. Patel¹ ¹Department of Robotics, Tsinghua University, Beijing, China ²Institute of Autonomous Systems, Warsaw University of Technology, Poland Zhu, T
Autonomous robots · Deep reinforcement learning · Task allocation · Modular navigation · Unstructured environments 1. Introduction Autonomous robots have transitioned from controlled laboratories to real-world applications: search and rescue, precision agriculture, and underground mining. However, three fundamental challenges persist: (i) partial observability in dynamic environments, (ii) coupling between low-level control and high-level mission planning, and (iii) sample inefficiency of monolithic learning approaches.
Autonomous Robots (Springer) Status: Submitted – Under Review (LetPub ID: AUTO-2026-0417) Abstract The deployment of autonomous robots in unstructured environments—such as disaster zones, dense forests, or planetary surfaces—requires robust navigation and real-time task allocation under uncertainty. This paper presents a novel modular framework that integrates deep reinforcement learning (DRL) with a dynamic graph-based task scheduler. Unlike end-to-end policies, our system separates perception (LiDAR + RGB), local path planning (SAC algorithm), and global task allocation (Hungarian algorithm with receding horizon). Experiments in both simulation (Habitat 2.0, Gazebo) and physical trials (Clearpath Jackal robots) show a 34% improvement in task completion rate and a 41% reduction in collision frequency compared to baseline DRL methods. Ablation studies confirm the modular design’s generalizability across unseen obstacle densities. We release the code and simulation environment for reproducibility.