October 20, 2021
Summary: Most Autonomous Mobile Robots (AMR) use teach-and-repeat learning that limits their effectiveness. Cloud Mapping unleashes true robot intelligence and maximizes facility throughput.
Have you ever heard the expression, “measure twice, cut once”? It’s better to do a little extra work now, and save yourself a lot of extra work later. This concept applies to autonomous mobile robot (AMR) installation in a very tangible way. When integrating these self-driving vehicles into your logistics workflows, you should take into account how different installation processes can affect their ability to function in dynamic, fast-paced environments.
Nearly everyone working in material handling can attest to the fact that industrial facilities are dynamic environments, with dynamic logistics workflows that can vary quarter to quarter, week to week, or even day to day. With recent trends showing increasingly rapid shifts in consumer demands, one never knows how a facility may have to adapt.
The teach and repeat is AMR programmed to follow specific routes cannot adjust to layout changes. As they only know how to copy – rather than how to adapt – they may be put out of service until remapped. This interruption wastes a good deal of time, one of the more valuable resources in many operations.
Right now, many warehouse robots are trained using “teach and repeat”. This method consists of:
Essentially, the AMR copies the route where it is driven, and is able to repeat the exact route once it’s shown. Seems easy, right? Well, it might be too easy. This method may get your robot up and running in no time, but you’ll be missing out on a lot of the benefits that a more in-depth installation process can provide.
Robots taught using teach and repeat can only remember individual routes, taking the same path from point A to point B. It will not be able to connect routes on its own, and every path selected must be chosen from a very limiting list of pre-programmed options. Unable to navigate outside its chosen path, the robot will need a little more guidance in decision-making. Someone will always have to choose what the robot does at each location, and it will need specific instructions and more oversight.
Perhaps one of the more costly drawbacks of this method is that the robot is often left to drive back from its job without any transports. Finishing one task only to return to start empty-handed is again, a waste of time and other valuable resources. The time this trip takes could be better spent, but Teach and Repeat AMRs cannot multi-task.
In contrast to Teach and Repeat, Vecna Robotics’ intelligent AMRs use onboard processing power and comprehensive maps to make decisions. Combined with path-planning technology, AMRs are not only able to multi-task, but can choose the most efficient way to accomplish their assigned task, avoiding traffic jams and factoring in multiple pallet assignments.
Vecna Robotics’ installation and deployment process involves mapping the entire facility once. Because the robots share information with one another, the mapping is done with one robot and then uploaded to the entire fleet, or any added robots later on – making fleet expansion easy. Instead of needing to remap robots every a route changes, the robots already knows every possible route. A fleet of Vecna Robotics’ AMRs can handle over a hundred assignments – and will figure out the best way to complete them. The robots answer to a Fleet Control Dashboard, managed by floor workers, in which tasks are assigned. Here, routes can be altered or “turned off” if necessary.
Rather than taking the time to remap – robots can be told to go anywhere at the click of a button. They are also able to intelligently navigate changing environments and dynamic logistics workflows using sensor data and advanced topological reasoning.
Learn more about the advanced navigation features of AMR technology here.