Papers
15
Total Citations
164
H-Index
8
About
Pritam Paral is a leading researcher in human-robot coexistence, specializing in vision and multi-sensor systems for human tracking and localization. His core contributions lie in developing photometric invariant algorithms for shoe detection—a critical sub-problem in leader-follower robotics—using techniques like DBSCAN clustering and OPTICS-based template matching, which have garnered over 30 and 22 citations, respectively. Paral’s work uniquely integrates chaos-based random sampling and fuzzy density clustering to ensure robust detection under challenging lighting and noise conditions. Beyond vision, he has advanced sonar-based human leg localization through Gaussian process regression and automatic relevance determination, achieving precise position estimation in dynamic environments. His recent research on rough entropy-based granular features and 2-D locality preserving projections addresses high-dimensional sensor data for robot navigation, with papers accumulating 18 citations since 2023. With a total of over 140 citations across his top ten works, Paral’s innovative fusion of manifold learning, density-based clustering, and probabilistic regression has significantly enhanced the reliability of human-robot interaction systems, making him a key figure in safe, autonomous coexistence.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10