ForwardX's Visual AMR Solution

Summary of Innovation

Our solution has 4 pillars:
1) Visual Autonomous Mobile Robot – Rather than relying on LiDAR SLAM like most competitors on the market, Forwardx’s robots depend on a sensor fusion solution with Global Award winning (footnote a) Computer Vision (CV) being the primary source for localization and obstacle avoidance on our robots which are fused with LiDAR, IMU, and Encoder data. LiDAR is 2D, so 3D CV technology offers a richer source of data, thus a more robust solution that is better suited to changing environments such as warehouses. This “robot visual fusion” is a new trend not only in material handling, but across autonomous vehicles, including Elon Musk’s war on LiDAR in favor of Computer Vision (footnote b).

2) Edge/Cloud based architecture – ForwardX takes advantage of edge/cloud based IT architecture. At the edge, our Robot Computing System (RCS) includes the multi-sensor fusion, computer vision based deep learning, local path planning ability based on V-SLAM and reinforcement learning. In the cloud resides our Robotic Process Manager (RPM). The RPM includes our multi-agent cluster scheduling system. It receives tasks from operators, WMS/MES/ERP, as well as the robots to holistically assign global tasks and intelligent paths for the robots to take, greatly increasing their efficiency.

3) Artificial Intelligence – ForwardX robots depend on V-SLAM with Deep Learning techniques for positioning and navigation. This allows the robot to adapt more readily to changing environments that LiDAR only based robots cannot cope with. We also “learn on the fly” as we attempt to predict velocity and direction of obstacles using Deep-Q-Learning and Asynchronous Advantage Actor-Critic techniques.. This creates a more intuitive nature in our robot leading to more collaborative and efficient local path planning.

4) ROI Less than 2 Years – Without an attractive ROI, it is difficult to mass deploy innovative technologies. ForwardX aims to keep the ROI for customers under 2 years for a full turn-key solution. We not only supply the technology, but also have a team of deployment engineers and project managers to ensure we deliver. In fact, we are so confident in our technology, that we will back agreed upon UPH improvements that are necessary for our customers to make an investment in AMRs.

Footnotes:
(a) Multi-object tracking – VOT-RT 2018: Number 1 in the world; IEEE’s multi-camera pedestrian detection and tracking competition (95% accuracy in Market 1501, 62% first hit rate in VIPER test set)

(b) (https://www.forbes.com/sites/bradtempleton/2019/05/06/elon-musks-war-on-lidar-who-is-right-and-why-do-they-think-that/#42e858812a3b)

Innovation Statement

ForwardX is the global leader in merging innovative technology from computer vision, artificial intelligence, and edge/cloud computing in a commercially viable package to the logistics, manufacturing, and retail markets. Robots can be deployed quickly and by integrating our RPM system with a customer’s WMS/ERP with our built in RFID solution, we are able to pick material more intelligently and accurately than any other existing solution on the market while avoiding people and other obstacles.

Applicability

Visual AMR solution with Deep Learning which will improve pick time and double UPH in any goods to person or person to goods application within ecommerce, manufacturing, and order fulfillment even in a retail environment that can be installed and running in one week, performance guaranteed, as well as a reduction in picking cost on average of 30%..

Posted Date	12/19/19
Contact	Nicholas Temple
Contact Email	nic.temple[at]forwardx.com
Category	Best New Innovation

« Back to Innovation Awards