Location errors of self-propelled machines and uncontrolled violations of protection zones generate costly delays in the timing of production lines. The choice of the appropriate navigation method determines the success of the implementation of AGV and AMR systems, directly affecting installation costs, reliability and the system's susceptibility to further expansion. This article analyzes the structure of infrastructure-free navigation, sensor data fusion algorithms and safety architecture consistent with current harmonized standards in MOBOT® mobile robots.

From rigid infrastructure to the flexibility of Free Navigation

Designing the movement trajectories of mobile robots has evolved from invasive methods that require floor modifications to full software autonomy. The choice of technology determines the accuracy parameters and operational flexibility in the structures of Industry 4.0.

Systems based on fixed infrastructure (AGV)

Classic AGVs move along clearly defined paths using inductive wires in the floor, magnetic tapes or optical lines.

• These systems ensure high repeatability and positioning accuracy of 2–5 mm.
• They are fully resistant to temporary or permanent changes in the surrounding geometry.
• They are used in low-profile trolleys that approach loads "from below" and in precise docking processes to assembly lines.
• The disadvantage of the solution is the high initial cost of infrastructure construction and low flexibility - each change in the factory layout requires physical installation work.

Laser Guided Vehicles (LGV)

LGV laser navigation uses scanners mounted on vehicle masts that cooperate with physical reflectors (triangulation mirrors) placed on the walls and columns of the hall. The vehicle determines its position by triangulating against known reference points. This method allows for work at higher speeds while maintaining high accuracy and no interference with the floor, but requires the installation of reflectors and the supplier's involvement in the reconfiguration of routes.

Natural navigation and SLAM (AMR) algorithms

The most advanced solution isnatural navigation (wolna nawigacja), w której robot rezygnuje z taśm czy luster, wykorzystując jako punkty odniesienia stałe cechy otoczenia: ściany, filary czy regały. Podstawowym komponentem jest technologia SLAM (Simultaneous Localization and Mapping).

In the startup phase, the robot performs mapping runs, generating a digital map of the surroundings based on a point cloud from LiDAR sensors or camera images. During operational operation, the robot controller constantly compares current measurements inscan-to-scan cycles oraz dopasowuje je do mapy bazowej (scan-to-map). Algorytm na bieżąco integruje te dane z odometrią (enkodery silników) oraz wskazaniami modułów IMU (Inertial Measurement Unit), co skutecznie eliminuje narastanie błędu położenia.

In advanced MOBOT® fleet systems, WObit implements a dynamic mapping mechanism, where the master map is shared by the entire fleet and updated with new, permanent elements of the environment.

Sensory data fusion and hybrid positioning

The industrial environment is characterized by high dynamics of change - pedestrian traffic, changing storage zones or moving forklifts may disrupt the operation of SLAM algorithms. To ensure location stability, WObit uses advanced data fusion and a hybrid approach.

Multi-level arrangement of sensors

The reliability of positioning and smooth movement of MOBOT® robots results from the simultaneous processing of signals from sensors mounted at different heights:

• Low-profile scanners: Placed close to the floor (often acting as safety scanners), they precisely identify obstacles in the critical zone immediately around the robot.
• Upper scanners / vision systems: They have a larger field of view and are used for stable location of the robot in relation to distant, unchanging structural elements of the hall.

Hybrid approach for critical accuracy

Although SLAM natural navigation significantly reduces installation time and costs, it may present localization difficulties in very rapidly changing environments. In projects with stringent dimensional tolerance requirements, WObit combinesnatural navigation z dodatkowymi referencjami. Zastosowanie fizycznych znaczników (np. kodów DataMatrix lub markerów podłogowych) w strefach dokowania pozwala robotom MOBOT® uzyskać powtarzalność pozycjonowania do 1 mm, stabilizując pobieranie i zdawanie ładunku.

External logistics: Integration of GNSS RTK systems

In applications combining internal logistics with transport in open areas (storage yards, interhall zones), the LiDAR laser signal does not find a sufficient number of reflection points from building walls. WObit solves this problem by integrating GNSS RTK modules (Global Navigation Satellite System with Real Time Kinematic correction) as another link in the MOBOT® architecture.

Thanks to the constant reception of corrections sent from the ground reference station, mobile robots moving outdoors achieve centimeter positioning accuracy (in the range of 1-2 cm). This allows you to precisely drive vehicles in designated external corridors and maintain safe distances from road infrastructure or buildings.

Smooth transfer of control: How do MOBOT® robots cope on the border between the hall and the maneuvering area?

One of the biggest challenges in autonomous inter-hall transport is the moment when the robot leaves the building. At the threshold of the hall, natural laser navigation begins to lose its previous reference points, such as walls or shelves. In turn, the precision satellite system (GNSS) in the gate area itself may encounter interference because the signal from the sky is reflected from the roof and facade of the building.

MOBOT® robots solve this problem using an intelligent data fusion algorithm that manages the smooth transfer of vehicle control. Instead of abruptly switching between driving modes, the robot's on-board computer evaluates in real time which sensors are more reliable in a given fraction of a second:

• Working inside the hall (Indoor): The robot is mainly based on laser scanning of the surroundings (SLAM) and a digital map of the building. This data is constantly verified by precise wheel rotation sensors and vehicle tilt detection systems.
• In the transition zone (entrance gate): As the robot approaches the exit, the system starts "listening" to the signal from satellites and corrections from the RTK ground reference station. The algorithm smoothly reduces the weight of the laser map, increasing the importance of satellite coordinates. Thanks to this, the risk of the so-called "position jumps", which in simpler devices could lead to location errors and a sudden, emergency stop of the robot in the gate opening.
• Outdoor work: After going fully outdoors, the GNSS RTK satellite navigation takes over the primary role, guiding the machine along the designated corridor with centimeter accuracy (in the range of 1-2 cm). At this time, laser scanners (LiDAR) change their function - they cease to be used for orientation in space, and focus solely on scanning the road to detect obstacles and protect people.

 

Integration with OT/IT structure and fleet traffic management

Autonomous navigation does not function in isolation - it requires close connection with the plant infrastructure, superior systems and building automation. MOBOT® robots exchange operational data via industrial Wi-Fi networks, industrial Ethernet or 5G networks. Master systems (WMS, MES, ERP) and PLC controllers receive real-time parameters regarding the robot's position, mission status and estimated time of arrival (ETA).

To maintain traffic flow, coupling navigation with building automation using the "corridor reservation" mechanism is critical:

• Access request: A robot approaching a chokepoint, intersection or fire gate sends a request to the master system.
• Priority assignment: The central fleet management system analyzes the position of other vehicles, assigns passage priority and blocks entry to other units, preventing traffic jams.
• Hardware control: The robot automatically generates a gate opening signal or forces a change in the state of industrial traffic lights. This solution works, among others, in a household appliances factory, where a robot system fully manages road traffic in the plant's external area by controlling traffic lights.

Industrial safety standards (PN-EN ISO 3691-4)

The task of MOBOT® robots is to take full control over the flow of goods while maintaining the highest standards of personnel safety. The design of the machines tightly integrates navigation data with safety circuits, ensuring full compliance with the PN-EN ISO 3691-4 and ISO 13849-1 (Performance Level d) standards.

Information about the current speed and steering angle of the vehicle dynamically modifies the range of protection zones in safety laser scanners. After detecting a person in the warning field, the vehicle slows down, and the violation of the critical zone triggers a hardware, two-channel power cut to the drives and the vehicle immediately stops. Processing signals from sensors mounted at different heights allows for the simultaneous implementation of two tasks: stable location based on distant objects and immediate response to low-profile obstacles or hanging infrastructure elements.

Summary – Hybrid navigation and full control over intralogistics

The implementation practice of WObit engineers clearly shows that in modern industry there is no single, universal navigation method that will work in every scenario. Maximum efficiency and stability of internal transport require flexibility and skillful combination of available technologies.

The transition from systems strictly dependent on infrastructure (such as magnetic tapes) to autonomous natural navigation and SLAM is a clear market trend that dramatically shortens the time and costs of installing robots in a plant. However, classic technologies and physical reference points still play a key role in places where the highest precision is a priority - for example in millimeter docking into production machines.

The introduction of advanced data fusion - combining laser space scanning, vision systems, odometry, and in outdoor applications also GNSS RTK satellite location - allows MOBOT® robots to meet the stringent safety requirements of the PN-EN ISO 3691-4 standard. This hardware and software architecture guarantees smooth machine movement in the dynamic Industry 4.0 environment, eliminating unpredictability, downtime and human errors.

By choosing a balanced, hybrid approach to navigation, the company can be sure that the logistics system will remain stable and repeatable around the clock, giving managers and production directors an absolute sense of control over the material transport process.

Engineering audit of navigation and safety Are you planning transport automation and want to precisely select a navigation system to suit the specificity of your halls and external routes? The WObit team of experts does not rely on guesswork - we support each technological decision with hard data and analysis of the work environment.

Contact our engineers to arrange a professional audit of the plant infrastructure for the implementation of a safe fleet of MOBOT® robots.