Design, Implementation, and Performance Optimization of a ROS Based Autonomous Mobile Robot for Intralogistics in Manufacturing Facilities

Figures & Tables

Parameter	Mean	Std. Dev.	Min.	Max.	Range
Tractional Force (N)	311.993	6.845	299.6	318.83	19.23
Total Driving Force (N)	74.388	37.38	19.38	128.42	109.04
Tractional Torque (Nm)	15.007	0.329	14.41	15.34	0.93

Feature	AMRs	AGVs
Navigation Technology	AI-driven sensor-based navigation (LiDAR, cameras, millimeter-wave sensing) [17]	Follow fixed paths using magnetic strips, beacons, or QR codes [18]
Path Dependency	No predefined paths; dynamically plans routes in real-time [17]	Fixed paths with minimal deviation from predefined routes [18]
Environmental Adaptability	Highly adaptable to unstructured and dynamic environments [17,19]	Limited to structured environments with predefined routes [18]
Obstacle Detection	Advanced AI-based obstacle detection with real-time path adjustments [20]	Basic obstacle detection; usually stops when encountering obstacles [18]
Operational Flexibility	High flexibility; can navigate new environments without pre-set guides [21]	Low flexibility; requires infrastructure modification for route changes [21]
Implementation Cost	Higher initial investment due to advanced sensing and AI [22]	Lower initial investment but higher cost for infrastructure setup [22]
Application Suitability	Smart factories, adaptive logistics, and warehouses [17,18]	Manufacturing lines, repetitive logistics, and controlled environments [17]
Path-Planning Algorithms	Hybrid A, RRT, D, and reinforcement learning-based methods [23]	Mostly rule-based or fixed path-following algorithms [23]

Hybrid A* Algorithm

Move Base (A* Algorithm)

Demonstrates higher precision in dynamic environments.
Continuously adjusts the path based on real-time sensor data.
Efficiently navigates around obstacles, ensuring collision-free movement.
Particularly effective in environments where frequent adjustments are necessary due to changing conditions.
Shows robustness in maintaining the robot’s pose with the aid of the AMCL algorithm.

Shows strong performance in real-time path adjustments and obstacle avoidance.

The robot’s planned path is dynamically updated based its position, as well as newly detected obstacles.
Maintains accurate localization and adjusts its trajectory efficiently.
Suitable for environments with well-defined obstacles and predictable changes.
Relies heavily on the AMCL algorithm for consistent and accurate pose estimation.

Slope Angle	Payload and System Mass (kg)	Tractional Force (F_tr) (N)	Total Driving (F_drv) (N)	Tractional Torque (T_r) (Nm)
0.0°	130	318.83	19.38	15.34
2.5°	130	318.52	33.28	15.32
5.0°	130	317.61	47.16	15.28
7.5°	130	316.10	60.99	15.20
10.0°	130	313.98	74.74	15.10
12.5°	130	311.27	88.38	14.97
15.0°	130	307.96	101.89	14.81
17.5°	130	304.07	115.25	14.63
20.0°	130	299.60	128.42	14.41