Yasir Salam

Mr. Yasir Salam


Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 January 2025 in Sensors
Verified
Journal Article
Open access
PDF file
YLYinbei Li
Yinbei Li
QLQingyang Lyu
Qingyang Lyu
JYJiaqiang Yang
Jiaqiang Yang
YSYasir Salam
Yasir Salam
BWBaixiang Wang
Baixiang Wang
Navigating crowded environments poses significant challenges for mobile robots, particularly as traditional Simultaneous Localization and Mapping (SLAM)-based methods often struggle with dynamic and unpredictable settings. This paper proposes a visual target-driven navigation method using self-attention enhanced deep reinforcement learning (DRL) to overcome these limitations. The navigation policy is developed based on the Twin-Delayed Deep Deterministic Policy Gradient (TD3) algorithm, enabling efficient obstacle avoidance and target pursuit. We utilize a single RGB-D camera with a limited field of view (FOV) for target detection and surrounding sensing, where environmental features are extracted from depth data via a convolutional neural network (CNN). A self-attention network (SAN) is employed to compensate for the limited FOV, enhancing the robot’s capability of searching for the target when it is temporarily lost. Experimental results show that our method achieves a higher success rate and shorter average target-reaching time in dynamic environments, while offering hardware simplicity, cost-effectiveness, and ease of deployment in real-world applications.

6 August 2025 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
YLYinbei Li
Yinbei Li
QLQingyang Lyu
Qingyang Lyu
JYJiaqiang Yang
Jiaqiang Yang
YSYasir Salam
Yasir Salam
WWWeiang Wang
Weiang Wang
Force-sensitive manipulation is essential for tasks such as cleaning, polishing, and surgical assistance, yet it remains challenging due to complex contact dynamics and the need for real-time adaptation. We propose DP-RRL, a hybrid learning framework that combines a diffusion policy (DP) for imitation learning (IL) with a residual reinforcement learning (RL) module to tackle these challenges. First, the DP leverages expert demonstrations to generate smooth, expert-like motion-force trajectories. Next, a residual RL component refines these trajectories online by adapting force commands to account for unmodeled contact dynamics. Our approach incorporates multimodal perception, including RGB point clouds, force feedback, and proprioceptive data, and achieves robust force-sensitive manipulation in unstructured environments. We validate DP-RRL on a 7-DOF robotic arm performing basin-cleaning tasks, where it demonstrates an average success rate of 88% across 12 unseen scenarios, outperforming state-of-the-art baselines.

4 November 2016 in The Journal of Engineering
Verified
Journal Article
Open access
PDF file
ANAli Nasir
Ali Nasir
YSYasir Salam
Yasir Salam
YSYasir Saleem
Yasir Saleem
The problem addressed in this study is that of a team of robots searching in an unknown area. The proposed solution is based on hierarchical agent architecture. Agents are formulated using Markov decision process model and search policy is calculated by solving the resulting Markov decision processes. In the proposed approach, one agent is the leader agent and remaining agents are member agents. Main duty of the leader agent is to assign the member agents to various partitions of the search space based on the information it receives. Member agents on the other hand, search in the assigned area until either the goals are achieved or their assignment is changed. There are certain advantages of the proposed approach over the existing approaches for multi‐agent search. For example, the computations within each agent have been limited to only one partition of the area at a time. Also the effective area to be explored by each agent is limited; therefore, less on‐board memory is required to keep track of how much of the search task has been completed. Furthermore, since Markov decision process (MDP) models are solved offline, the online computational requirement is reduced.

15 April 2025 in Autonomous Robots
Verified
Journal Article
Closed access
PDF file
YSYasir Salam
Yasir Salam
YLYinbei Li
Yinbei Li
JHJonas Herzog
Jonas Herzog
JYJiaqiang Yang
Jiaqiang Yang

26 December 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
YLYinbei Li
Yinbei Li
QLQingyang Lyu
Qingyang Lyu
JYJiaqiang Yang
Jiaqiang Yang
YSYasir Salam
Yasir Salam
BWBaixiang Wang
Baixiang Wang
Navigating crowded environments poses significant challenges for mobile robots, particularly as traditional Simultaneous Localization and Mapping (SLAM)-based methods often struggle with dynamic and unpredictable settings. This paper proposes a visual target-driven navigation method using self-attention enhanced deep reinforcement learning (DRL) to overcome these limitations. The navigation policy is developed based on the Twin-Delayed Deep Deterministic Policy Gradient (TD3) algorithm, enabling efficient obstacle avoidance and target pursuit. We utilize a single RGB-D camera with a limited field of view (FOV) for target detection and surrounding sensing, where environmental features are extracted from depth data via a convolutional neural network (CNN). A self-attention network (SAN) is employed to compensate for the limited FOV, enhancing the robot’s capability of searching for the target when it is temporarily lost. Experimental results show that our method achieves a higher success rate and shorter average target-reaching time in dynamic environments, while offering hardware simplicity, cost-effectiveness, and ease of deployment in real-world applications.

6 November 2025 in IEEE Access
Verified
Journal Article
Open access
YLYinbei Li
Yinbei Li
QLQingyang Lyu
Qingyang Lyu
YSYasir Salam
Yasir Salam
JYJiaqiang Yang
Jiaqiang Yang
RNRafik Ighil Nessouk
Rafik Ighil Nessouk
Public restroom cleaning is a laborious but critical task for public hygiene, offering both an opportunity and a challenge for robotic automation. Traditional robotic systems struggle in such environments due to partial observability, dynamic conditions, and reliance on rigid task scripts and brittle perception pipelines. This paper presents a hierarchical task planning and skill grounding framework leveraging large language models (LLMs) for autonomous restroom cleaning using a mobile manipulator. A high-level LLM (LLM-H) generates semantic task plans based on symbolic scene representations derived from a visionlanguage model (VLM), while a low-level LLM (LLM-L) refines these plans into a sequence of executable, parameterized robotic skills. To ensure robust and adaptive operation, the framework integrates a semantic active perception module that mitigates visual occlusions through targeted exploration, a dynamic task buffer that reprioritizes tasks in real time to handle environmental changes, and a VLM-based validation module that assesses task outcomes and triggers corrective actions when necessary. Experiments conducted in real-world public restrooms demonstrate that our system outperforms state-of-the-art baselines in task-completion rate while robustly handling partial observability, dynamic events, and long-horizon cleaning sequences. Its modular, scalable approach provides a solid foundation for deploying intelligent service robots in hygiene-critical public spaces.