Bannière XuP-Host

What are the characteristics of an Autonomous Mobile Robot?

The Mobile Cobots offered by Meanwhile are totally autonomous in their journey. To do this, these robots use Artificial Intelligence (called SLAM: Simultaneous Localization And Mapping) allowing them both to locate themselves at any time and to dynamically avoid any obstacle that has not been mapped.

Thus, Mobile Cobots are able to navigate in complex environments, with many operators and handling equipment, while ensuring optimal efficiency in their transport. Users can then refocus on their higher value-added tasks while the Mobile Cobots take care of transporting the products from one station to another.

I. What is an autonomous mobile robot?

When we talk about intelligent mobile robots, we are talking about AMR (Autonomous Mobile Robot). The AMR belongs to the new generation of automatic trucks called AGVs (Autonomous Guided Vehicles). The main particularity of AGVs lies in the fact that they are guided by a rail (real or virtual). Their location is done using infrastructures such as beacons posted on the walls, in the ground, guide rails or location studs. The certain advantage that can be conferred on them is the determinism of their course. In normal operation, they will always take the same amount of time to cover their traced path between two points. However, in complex, reconfigurable and dynamic environments, this technology quickly reaches its limits, and its commissioning becomes very expensive.

The AMR, on the other hand, scans its environment using a previously learned map and its many location and security devices. It does not need to be guided which minimizes the installation effort. It is also able, depending on the settings applied, to optimize its route and to circumvent obstacles in its path or quite simply to decide to take a different non-predefined path.

II. What are the particularities of autonomous mobile robots?

Meanwhile robots are equipped with artificial intelligence specialized in indoor navigation (SLAM). SLAM, Simultaneous Localization And Mapping, allows the robot to construct its environment and modify its behavior according to unmapped obstacles while locating itself in real time. In order to move independently and without a predefined trajectory, the robot will combine its own information with information from its environment (returned by its lasers and sensors.)

To be considered as such, an AMR must meet the following characteristics:

  • Movement planning

To carry out its mission, the AMR has the ability to optimize its route. Its starting configuration will define its global trajectory and will evolve according to the data retrieved in real time, which will alter its local trajectory. The AMR will thus continuously calculate its collision-free path. In other words, to carry out its mission, the Meanwhile robot optimizes its route, which it defined itself at the start (global trajectory), while avoiding any obstacles that may come its way (local trajectory).

  • Localization

To be able to plan their journey independently, Meanwhile mobile robots are able to locate themselves in their environment. They use a number of on-board sensors, such as laser scanners, vertical lasers etc. To be permanently located, the Meanwhile robots will, in real time, compare their mapping to the environment modeled by their sensors. Thus, any intrinsic positioning fault is instantly corrected and the robot is permanently localized. Moreover, when the work environments are very dynamic, it is possible to improve the localization of the robot by a light triangulation system, called Acuity.

  • Natural navigation

The mobile robots are able to calculate the commands of the actuators of the system allowing to carry out the planned movement. The robot plans its own trajectory. It follows the planned path (global trajectory) while avoiding unmapped obstacles on the way. If the robot encounters obstacles, it modifies its “local” trajectory to avoid them, but if the latter cannot continue (for example, a door is closed), it globally replans its path (for example, to find a different door.)

  • Reactive obstacle avoidance

In addition to the main safety laser scanner, located 200mm from the ground, Meanwhile mobile cobots are equipped with various sensors allowing them to circumvent and avoid obstacles:

A “foot detection” laser scanner located 60mm from the ground. This sensor allows the Cobot Mobile to detect and circumvent any obstacle located at ground level.

A row of sonars located on the back of the Cobot Mobile. These sensors allow a safe backward movement of the Cobot Mobile. Note that the Cobot Mobile only reverses in identifiable areas and never on its own initiative (calling a function during a sequence) (approaching the charging station for example).

Vertical laser scanners, allowing the creation of two planes on either side of the Cobot Mobile. It can then reliably detect any overhanging object located in a different plane from the horizontal scanners (forks of forklifts, operator tables, etc.) as well as any slope in the ground. This data is used during navigation and allows the Cobot Mobile to move optimally between obstacles in complex environments.

  • Rapid deployment

The installation of the Meanwhile autonomous mobile robots is carried out in extremely short times because it is not necessary to modify the existing infrastructure. Just as the robots can be reprogrammed easily, if the infrastructure and the need evolve. (Our customers are able to modify the delivery points themselves.)

The deployment of the Meanwhile mobile robots takes place in 3 steps: