Welcome to MRDVS Hub
Camera Hardware
Industrial Solutions
User Cases
Frequently Asked Questions
Ecological Partners
Top-view navigation solves the 3000+ warehouse positioning and navigation challenges for unmanned forklifts.b
type
Page
status
Published
date
slug
summary
tags
category
icon
password
The article originates from Gaogong Robotics, authored by Zhang Haocheng.
With the advancement of LiDAR technology, laser navigation has now become one of the mainstream Navigation Mode in the industry. AMR solutions based on laser SLAM navigation have been implemented in numerous scenarios.
As the application of laser navigation technology increases, its shortcomings have also been exposed, particularly in the stability of positioning, which struggles to withstand the challenges of complex scenarios.
In complex and ever-changing environments, the location of goods is constantly shifting, with few fixed reference points. For AMR （Autonomous Mobile Robot） using laser SLAM solutions, frequent entry and exit from storage, along with constantly changing positions, mean that the contour environment scanned by the laser is always changing. Achieving reliable localization in such a continuously changing environment is extremely challenging.
In large, open workshops with widely spaced steel structure columns, the reliability of laser SLAM navigation for AMR （Autonomous Mobile Robot） diminishes in such environments due to sparse laser-scanned point clouds and a lack of effective positioning references.
Visual SLAM navigation captures surrounding images through a depth camera mounted on the robot and generates dense point clouds maps. The collected environmental information is sufficiently rich, and local changes in the environment do not affect the robot's positioning, thus providing extremely high scene adaptability.
Take the "difficult path"
Since visual SLAM technology holds an unparalleled advantage over laser SLAM in terms of environmental adaptability, why are there so few robotics companies in the market that actually adopt visual SLAM?
The primary reason affecting the choice of visual SLAM for AMR （Autonomous Mobile Robot） is the high technical barrier of depth vision technology. Most AMR companies lack the capability to independently develop 3D vision function modules, as they neither have a visual R&D team nor the relevant technical expertise.
MRDVS, though a young company, boasts core technical members with nearly two decades of experience in optics, computer vision, and other fields. The company possesses comprehensive R&D capabilities in computer vision and AMR （Autonomous Mobile Robot）.
Compared to mature LiDAR technology, depth vision can be considered a "more challenging path," and MRDVS has chosen the latter.
After years of technological accumulation and breakthroughs, MRDVS has established a robust and comprehensive Mobile Robot Depth Vision System (MRDVS), which includes self-developed 3D vision sensors and depth vision perception algorithms. This makes the company the first in China to provide an integrated 3D vision hardware and software solution for AMR （Autonomous Mobile Robot）.
By integrating MRDVS, AMR （Autonomous Mobile Robot） can achieve visual positioning, 3D Vision Obstacle Avoidance, and high-precision docking, with significantly improved safety, stability, and intelligence, meeting the demands of more complex application scenarios.
Taking the warehousing scenario of a large domestic contract manufacturing enterprise as an example. It is understood that the enterprise has a floor-stacked warehouse with over 3,000 storage locations, where goods are frequently moved in and out. The pallets and goods are constantly rearranged, with no fixed reference points. Additionally, during operation, robots may encounter workers and cardboard boxes left in the aisles. At the same time, for the overall aesthetics of the warehouse and the convenience of deployment, the customer does not want to paste 2D Code Camera on the floor.
In this scenario, the company has limited options, with only laser SLAM and visual SLAM unmanned forklifts available. However, for AMR （Autonomous Mobile Robot） using laser SLAM, achieving reliable localization in constantly changing environments is extremely challenging. In the end, the company chose Lanxin Technology's visual SLAM unmanned forklift solution.
It is understood that MRDVS has provided Lanxin Technology's unmanned forklifts with an integrated solution for visual positioning, pallet docking, and visual obstacle avoidance.
In terms of positioning and navigation, the forklift employs a 3D visual SLAM solution. Using a depth camera to capture unmarked three-dimensional environmental information, a dense point clouds map is generated. The collected data is linked to the robot's actual position to achieve autonomous robot localization and navigation. The advantages include: 1. Reliability and stability, unaffected by people, vehicles, or logistics; 2. Adaptability to significant scene changes, unaffected by indoor layout adjustments or movement of goods; 3. High precision at target points, typically within +/-1cm in general environments.
In tray docking, this forklift employs a 3D vision docking solution. By using a self-developed docking camera to capture images and generate a point cloud map of the pallet, combined with Lanxin's visual perception algorithm, the system calculates the pallet's deviation value, guiding the forklift to automatically adjust its position and fork direction, successfully picking up the pallet.
In addition, the forklift is equipped with a 3D Vision Obstacle Avoidance system, which uses obstacle detection cameras to identify overhead and low-lying obstacles within the field of view, ensuring the safety of the robot's movement.
View more cases
🏅Top-view navigation solves the 3000+ warehouse positioning and navigation challenges for unmanned forklifts.Top-view navigation solves the 3000+ warehouse positioning and navigation challenges for unmanned forklifts.Loading...
