Why?

The industrial trend is open, fenceless robot systems to allow easy access for humans to the robot cell, to assist and collaborate when desired. However, this poses challenges related to safety, human comfort and operation functionality.

Safety Monitoring System (SAM) offers this functionality by utilizing state-of-the-art object sensing technology to develop a complete low-cost, add-on system with cutting-edge functions for human safety, comfort and operation.

What?

SAM is a low cost, advanced safety monitoring system providing safe, fenceless, flexible operation of industrial robot systems. It can cover a single robot cell or a factory.

  • SAM Detector = Radar and Vision-based solution for protection zone monitoring and motion tracking
  • SAM Processor = A software solution to a robot cell controller processing the detection of related Safety and Comfort incidents
  • SAM Bridge = A on-the-fly production scheduling solution synchronizing robot cell controllers and Manufacturing Executive Systems

In Brief

  • SAM4ROB is a research project, funded by The Eurostars Programme; offering a Safe Area Monitoring (SAM) system for fenceless factory automation.
  • SAM provides a natural, safe, synergic coexistence and collaboration between humans and robots even in harsh industrial environment
  • SAM4ROB Consortium :
    - 4 SMEs (PPM Robotics SA, YUJIN Mechatronics, DLM Solutions and F&P Robotics AG)
    - 2 Research Institutions (DGIST, BME)

Safety approaches

Proactive Comfort Measures in Social zone

Establish a natural behavior of the human to comply with risk avoidance. Provided by upfront transfer of information / knowledge about the potential safety risks Activated by a Social Zone, located outside the physical system area.

Proactive Comfort functions

  • Transfer of knowledge the robot's properties and behavior, to give the human an upfront understanding of when / where the robot is safe and dangerous.
  • The knowledge is transferred initially to the human (to new persons); thus, BEFORE the human enters any working situation that involves interaction with the robot / robot cell
  • The knowledge transfer is done by demonstrations

Reactive Comfort Measures in Notification zone

Provide a Notification to the human about a safety measure near to being activated if the Notification line is crossed.

Reactive Comfort functions

  • When the Comfort Transition Zone is crossed, a short notification is given to the human
  • Type of notification depends on the motion direction of the human; thus, whether the danger is increased or decreased
  • Comfort also includes measures to increase the physical and mental comfort when accepted safety conflicts occur

Proactive Safety Measures in Warning zone

Upfront measure, guaranteeing to reduce the consequences of any near to happen, safety conflict ; Activated by one or more Warning Zone, outside the Danger Zone and immediately, when a potential safety conflict is unlikely to be avoided.

Proactive Safety functions

  • The safety measures are zone based, thus, adapted to the risk for injuries
  • Within each zone, the following measures can be done
    • Constrain the maximum robot speed is adjusted, to guarantee the robot to stop before the human reaches the Warning Zone.
    • Postpone the task, and move the robot into safe stop, so the human can safely enter the WZ
    • Reconfigure WZ, to ensure longer safety distance to the robot
    • Reconfigure WZ, and choose another task / other path, which the robot can perform normally even within the limited WZ (But no task sharing between the human and the robot)
    • Switch ownership to a working area shared between the human and the robot (to be operated in shared mode)
    • Switching the robot into safe operation mode (safe speed; to be operated in cooperation mode)
    • Switching the robot into collaboration mode (safe speed, soft robot; to be operated in collaboration mode)

Reactive Safety Measures in Danger zone

Immediate measure, guaranteeing protection against injuries when a safety conflict occurs; Safeguarded passively, by physical safety features inside the Danger Zone or actively, by enabling safe mode in the control system, when the human is crossing the Danger Zone.

Reactive Safety functions

  • Within the Danger Zone, the robot is performing its task, and thus, might collide with objects / humans being present
  • Safety measure MUST be implemented to prevent any injuries of humans
  • Safety measures are activated WHEN the human is inside a Danger Zone

Modules

Three distinct products with their own dependencies
In line with the modules discussed into application

Functionalities

Reactive Safety
Actions to avoid the risk for / any consequences of a safety conflict when the human is crossing the Danger Zone:
Proactive Safety
Actions to reduce the risk-for / consequences-of a safety conflict when the human is crossing the Warning Zone:
Reactive Comfort
Forewarnings to avoid safety actions to be enabled with a notification when the human is crossing the Notification Zone:
Reactive Comfort
Cognitive actions regulating the interaction between the robot and the human when the human is crossing the Social Zone:
Flow
On-the-fly actions controlled by a MES on the factory level based upon the human's position inside the factory:

Features

Reduced floor space

With long range detection of moving objects we discover potential conflicts early, and thus, don't require the normal, extra safety margins for robot cells.

Increased flexibility

On demand configuration of the robots' safety zones and safety modes.

On demand switching between automatic, co-located and collaborative mode

Increased comfort

Notification about potential conflicts can reduce the stress caused by unnecessary emergency stops. Robots have “social skills” to make them less scary.

Increased productivity

Open robot systems can work at full speed, when people are not in danger.

Proactive conflict solving increases the up-time of the robot cells.

Increased comfort and productivity by proactive means
Early moving object detection
  • Predict objects’ intended action
  • Distinguish between humans and inanimate objects
Proactive conflict avoidance
  • Determine potential conflicts
  • Conflict reduction with pre-notification of the user
Humans and robots in harmony
  • Robots with artificial, social skills
  • Automatic task selection for co-existing and collaborative work
SAM4ROB Consortium
Prof Trygve Thomessen, Project Coordinator
trygve.thomessen@ppm.no
+47 92 24 21 89

Contact