AgiBot Releases Genie G2 Max: Dual-Arm Rated Payload of 40kg, Taking on Heavy Work on the Factory Floor

AgiBot releases Genie G2 Max: combining heavy payload and fine manipulation, robots enter factories to take on heavy work

Xiao Lin 3 hours ago

AI capabilities are landing on production lines.

On October 26, AgiBot officially released the safety-grade heavy-duty embodied robot Genie G2 Max.

This product is mainly aimed at industrial scenarios such as depalletizing and palletizing, material transfer, and production line feeding. Its single-arm payload reaches 20kg, dual-arm rated payload is 40kg, peak payload can reach 50kg, and arm repeat positioning accuracy is within ±0.5mm.

Image source: AgiBot

Judging from the specifications, the G2 Max must both carry heavy objects and complete fine operations. At the same time, AgiBot is also trying to make this robot adapt to factories' existing production environments, minimize the cost of production line modifications, and enable the robot to be put into use faster.

This also makes the G2 Max's highlights go beyond payload capacity itself.

Let's start with the most direct handling tasks. In factories, material transfer may seem simple, but in reality there are many details to handle. A material box may contain dozens of kilograms of parts; the robot must have enough strength to lift it, control its posture during movement, and finally place the box accurately in the designated position.

The G2 Max uses 7-DOF force-controlled dual arms. Combined with visual guidance and dual-arm force control technology, it can handle various industrial materials such as cardboard boxes, Euro-standard turnover boxes, and irregular parts, reducing the workload of repeatedly teaching the robot for different materials.

On the production line of a leading domestic daily chemical FMCG company, the G2 Max has already been applied to mixed-line incoming material scenarios. Facing material boxes of different sizes and weights, it achieved a 100% material recognition rate and completed a series of operations including depalletizing, box opening and feeding, box film covering, and empty box recovery.

This means that the robot's work is no longer limited to a single transfer, but has begun to attempt to complete multiple tasks coherently. For production lines that need to handle multiple materials, this capability helps reduce manual intervention between different processes.

Image source: AgiBot

In addition to payload and precision, whether a robot can adapt to existing sites is also an unavoidable issue for enterprises when introducing automation equipment.

Especially for factories and warehouses already in operation, if introducing new equipment requires re-planning workstations and modifying production lines, deployment costs can easily become a threshold.

The G2 Max takes such needs into account in its design. Its vertical working range is 0.1 meters to 2.5 meters, its horizontal working coverage is 1.2 meters, and it is equipped with a high-precision waist pitch joint, allowing it to complete single-side depalletizing and palletizing by adjusting its body posture.

As a result, pallets can be placed directly against the wall, with no need to reserve an extra 1.5 meters of clearance space behind them, and low-level picking and high-level palletizing can also be completed by the same device.

In a leading e-commerce warehousing scenario, the G2 Max has been able to connect with the stereoscopic warehouse's "goods-to-person" system and work with existing robots to complete tasks such as full-box depalletizing and warehousing, and empty material box palletizing and recovery. According to AgiBot, its overall operation success rate exceeds 99%, and it can depalletize 80 boxes and palletize 75 boxes per hour.

For warehouses and factories, the efficiency of automation equipment is certainly important, but whether it requires less modification and can be deployed quickly also affects whether enterprises are willing to introduce it. If a robot can adapt to the original environment, it has the opportunity to cover more practical scenarios.

On the Qiyuan T1 production line, the G2 Max also handles material transport for the joint assembly line, responsible for delivering parts such as shoulder joint housings and hip joint assemblies from the material stacking area to the final assembly workstation.

These parts have high requirements for handling precision. Once the mounting surface is bumped, it may affect subsequent assembly. In the past, this type of work mainly relied on manual box stacking, carts, and loading and unloading. After the G2 Max was put into use, it can carry material boxes with both arms and complete material placement with sub-millimeter precision.

In other words, robots have begun to participate in the production process of another robot.

At the same time, the production line also has strict requirements for work rhythm. The production takt of the Qiyuan T1 is one unit off the line every 2.5 minutes, which means material supply must keep up with production progress. Zhou Ye, factory director of the robot manufacturing plant at Huguang Co., Ltd., said that the G2 Max's work takt is relatively stable, and its all-day takt curve is almost a straight line.

Image source: AgiBot

To meet long-duration operation needs, the G2 Max is equipped with dual-battery hot-swapping and autonomous recharging solutions, supporting 7×24-hour uninterrupted operation. In terms of safety, it integrates 2D/3D LiDAR, ultrasound, chassis safety edges, and physical emergency stops to build a 360° multimodal perception protection system.

In addition, the G2 Max is also compatible with the Genie Studio 2.0 embodied operation platform, supporting task orchestration, simulation verification, and cluster operations and maintenance, helping customers shorten on-site deployment cycles.

It must be said that competition in embodied intelligence has entered a more specific stage. Models and algorithms are still important, but hardware design, force control capabilities, environmental perception, safety protection, and coordination with existing production systems have all begun to affect how far a product can go.

No matter how outstanding a single capability is, if it cannot be integrated into a factory's workflow, it will still be difficult to truly achieve large-scale application.

And the reason industrial scenarios deserve attention is that they provide a relatively clear set of testing standards. Of course, robots entering factories does not mean that all positions can quickly achieve automation. The environments, materials, and production processes of different factories vary greatly. To make robots adapt to these changes, continuous technological iteration and on-site accumulation are still needed.

Image in this article from: 123RF licensed image library Infringement report Return Share Related articles All comments