Faster Robot Programming - AUTOMAPPPS news and press

2024-12

New Integration Partner: Accell GmbH became silver-level integration partner - based on experience after training several staff members in using AUTOMAPPPS software and successfull completing projects with above threashold volume - which are using AUTOMAPPPS planning software and computer vision.
ACCELL is specialized on reactive systems from small to large. ACCELL - with focus on easy to use small and medium-size systems to manufacturers and as subsystem provider to line-builders. ACCELL, has a build its reputation for providing easy-to use and even one-button robot-cells especially in grinding and sanding in the steel, construction, wood and automotive and other high-tech sectors.

2022-05

Platinum Level Ingegration Partner: Asis GmbH became our first integration partner on platinum level - which considers experience in terms of (number of) robot experts trained on AUTOMAPPPS software and (volume of) projects - based on AUTOMAPPPS software - successfully completed. ASIS has become a full service partner for reactive robot programming with AUTOMAPPPS and robot off-line programming (OLP, OLRP) - and has buit up a strong interdisciplinary team of surface, robotics, PLC and software experts that has been trained by us in depth in all aspects of AUTOMAPPPS paint-repair to provide you with the best service for reactive robot systems and real-time robotic-applications.

2020-04

Industrial state-x-time planning reduced robots´downtimes by more than 95%

Technical Background: “State-x-time planning” as method in academic R&D has been invented around 15 years ago for robots operating in environments that change over time (see e.g. S. LaValle: „Planning Algorithms“, 2006). We will shortly describe the details and the impact it has on industrial robotics.

Traditional robot motion planning does plan the robot´s motion in the state-space C(x) with x being the state of the robot. This was extended to include the time: C(t, x(t)). A full kino-dynamic planner extends the state space to C(t, x(t), dx(t)/dt) or C(t, x(t), dx(t)/dt, d2x(t)/dt2), i.e. represents not only the joints’ state, but also their velocity and possibly the acceleration. „Space-x-time“ extend the state x to include the changes on the robot cell over time as an additional state.

Convergent IT started more than 10 years ago to transfer state-x-time from its original researched applications such as mobile service robotics to industrial robots, which has to deal with less changes in the environments, but with more complexity by means of more joints - and even tougher: more constraints originating from the process to automate. As an illustration: the state-x-time planer AUTOMAPPPS does not plan e.g. a 7 axes robot in a state-space C of dimension 7, but in a 15 dimensional state space: 2 dimensions per joint (x(t),dx(t)/dt) plus one dimension for the changes of the robot cell over time. Additionally, all joints are speed- and acceleration-bound which influence the planning. For a cell with 2 robots, the dimension of the state space C is then 29, 4 robots 57 a.s.o..

Impact: Usage of this dry mathematics is rewarded with great opportunities. Not only novel very high-end applications, such as self-programming robot cells in the field of autonomous spot-repair benefit, because they wouldn´t be possible without real-time programming of robots in seconds by AUTOMAPPPS state-x-time with line-tracking support. Also others, currently more common robotic tasks, which could be programmed otherwise, do benefit: The method does allow huge cost-reduction and increases of OEE (overall equipment efficiency). One of the examples is shown in the following video recently permitted for disclosure.

The example depicts the painting of complex shaped parts, as common e.g. in investment goods or in machine builder industry. One of the key challenges is, that large robots have to enter narrow areas with a clearance or safety distance between robots and parts or chains being only a few cm. The robots are painting parts from the inside and or between chains – while the parts are moving.

Where other SW-tool has not been useable, AUTOMAPPPS has been able to support a save off-line programming efficiently. Time-consuming robot teaching with plenty of try-and-error and iterations is no longer needed. Thus, the state-x-time planning is reducing the non-productive time of the robot cells for more than 95% and beyond. What remains is the time needed to measure the location of one part per new class. Also, programming these challenging tasks can be done by the staff of the manufacturer - not by robot experts only. Those cases are illustrative examples for applying R&D methods in other large robotic domains. Maybe not as spectacular on the first view as its application in automatic car repair, but impressing after taking a closer look - and with a big advantage for the user.