Mixed Reality – About

Posted on Posted in Mixed Reality_en

The key feature of the Mixed Reality Competition is the combination of virtual and physical elements. Real robots (micro robots) with a floor area of about 2 cm² and about 2 cm height play on a virtual playing field with a virtual ball.

On the one hand, movements are generated by different speeds of the two wheels; on the other hand, the virtual ball is indirectly steered by means of an instruction. These circumstances require a meaningful linking of two different concepts:

The discrete behavior of a virtual world must be brought into harmony with the complex physical reality. The mixed reality strikes a bridge between simulation and reality, in order to connect the leagues. In the third year after the founding of the Mixed Reality, the games are played against 5. First tests 7 against 7 are carried out in order to be able to play with the future third system 11 against 11 in a scale-appropriate field.

 

Mixed Reality – Focus on Development

Posted on Posted in Mixed Reality_en

In the field of mixed reality, we have focused on research and development.

Vision Tracking

Vision tracking is of particular importance in the area of mixed reality. The goal of the mixed reality is to play 11 against 11. However, since the system is still designed for 5 to 5, a lot has to be improved.

Behavoir Management

Since the lower functions of the artificial intelligence already work very well, we are currently working on behavoirs, ie a roll-based system in which the robots dynamically change their roles. In this way, we hope for a better interaction between the robots and improved behavior.

Reinforced Learning

We also use this area of artificial intelligence and try to expand it. The advantage with self-learning algorithms is that although you do not know much about a system, you can still achieve good results.

Systemoptimierung

There are other ways to improve the mixed reality system. In order for our goal to be achieved by 11 players per side, a lot must also happen here. This includes, among other things, the development of more durable robots, as well as a better infrared.

Humanoid KI-Simulation

In robotics it is common to work with simulators. This saves time and money. We use the mixed reality system to test and develop our artificial intelligence for humanoids.

Mixed Reality – System

Posted on Posted in Mixed Reality_en

system2The Mixed Reality System is a unified system of real physical and fully simulated parts. When you see a mixed reality system for the first time, the big display is the first to appear. Even if the display has no influence on the game, it serves on the one hand as a playing field, as well as to give the viewers an insight into the simulated part of the world. The robots (Eco! BE2008) are responsible for the physical part of the system. In order for the robots to be able to interact with the simulated ball, their position must be determined. Each robot has a unique marker on its head. A camera films the playing field and the obtained image data are analyzed. The position of the markers is determined in the images. These are sent to the server. The server simulates not only the ball, but also all interactions with the ball, e.g. The collision between the robot and the ball. The positions of each robot and the ball are passed on to the clients by the server. The clients can use this data to send commands to the server. The commands are executed by the server directly, or they are sent via infrared to the robots.

@Work League

Posted on Posted in YouBot-en

The RoboCup @ Work League started in 2012. The aim of this league is to develop an autonomous robot that is to be used in the industrial environment and to handle complex tasks there. These tasks range from pure navigation within the arena, which in this league, e.g. A workshop simulates the interaction with objects up to complex co-operative or competitive problem solving. This league is mainly about the professional use of the different techniques, so tape and hot glue in this league are no alternative. This fact provides for a better planning of the different installation and conversion measures. In principle, it is possible to use any robot in this league. However, we have chosen Kuka’s YouBot to be a working and tested platform where we can advance our development efforts.

@Work focus

Posted on Posted in YouBot-en

Simulation

For testing new software and for machine learning, it is important to have a simulator that simulates the behavior of the real robot as accurately as possible. This does not stop with the mere movements, so also the kind of the control must be modeled after the original down to the smallest detail.
Localization and navigation

Navigation

is an enormously important field in robotics. Without the knowledge of the current position, the search for a way to a goal is already doomed to failure. It is therefore important to recognize the current position in the room as quickly and as accurately as possible. It is then possible to plan a route to the destination that promises the greatest economy of movement. The omnidirectional platform of the YouBot creates completely new possibilities for a goal. This makes it possible to achieve the desired orientation during the journey and thus to save time during positioning at the target.
manipulation

Manipulating objects

presents a great challenge. Since it is not possible to collect exact data through the various sensors, the localization of objects in space is based on pure assumptions, with the robot opting for the most probable position. Addressing these inaccuracies in the 3-dimensional space is a great area of ​​manipulation. The other large area concerns the arm control. It is important that the solutions of the inverse kinematics have to fulfill certain conditions. Thus the arm must not move into obstacles or into itself during the movement. It is also necessary to find out from which positions the arm can reach the object at all.
Artificial Intelligence

In order to achieve an autonomous action of our robot, it is inevitable to realize an artificial intelligence, which takes care of the decision-making based on sensory data and tasks. As in all other areas of robotics, we must also develop an approach that can deal with an inaccurate or incorrect picture of the environment and can bring this picture of the environment back into a consistent state. Our approach to artificial intelligence uses the SMACH project developed within ROS. This is an approach based on a state machine. In addition to the debugging and visualization capabilities provided by SMACH, the state-machine approach also offers many advantages that lead to a clearer and more consistent behavior of artificial intelligence.

Humanoid robots – Focus of development

Posted on Posted in Humanoid-en

The humanoid robots must bring many skills to participate in the football matches. Motion sequences like walking on two legs or shooting are great challenges for a robot. But also the processing of the image data and interpretation of these are the focus of the development. In addition to the software is always worked again and again to improve the hardware.


To run

The human-like running of robots is not very easy. Thus man needs the big toe round his foot to create a dynamic movement. In the case of robots, this is often attempted with a “tap step”. We are currently working on a modified version of the Walkengine of the DARwin-OP, which creates a stable running motion.


To shoot

In football, after shooting, shooting is of similar importance and therefore also one of our priorities. Here, not only as wide a range as possible and precise meeting are required, but above all stability. As one of the next goals, we want to reach a university.


Image processing

Image processing is one of our key areas. Here we are working on optimizing the algorithms and increasing the detection rate. The more robust the image processing, the better the behavior of the humanoids can be controlled.


Localization

As a further focus, we are currently dealing with localization, that is, The ability of the robots to determine their own position in the playing field. Self-localization is the first step towards a more abstract artificial intelligence.

RoboCup Germany 2016

Posted on Posted in News-en

This year, too, the WF Wolves RoboCup team from Ostfalia will compete at the World Championship for Robotics. This year the World Cup in Leipzig takes place from the 30.06. To 03.07.2016 instead. The team of the Ostfalia is in the categories RoboCup @ Work and Humanoid League (TeenSize). As in the past year, the students are working with their Brazilian colleagues at the Universidade Federal de Santa Maria (UFSM).

The current Leipzig team consists of 14 members. Five team members come from the Brazilian university. The WF Wolves are supported by professors Reinhard Gerndt (Ostfalia) and Rodrigo Guerra (UFSM) as a student project group.

Already since Monday, the students work to find the optimal setting of the robots for the conditions.

The RoboCup @ Work League:

The @work league is an industrial league. On the other hand, autonomous platforms, on which a robot arm is located, must successfully meet certain challenges. These challenges consist essentially of mastering a pre-defined course. As a particular challenge, its details can be varied in the course of the process. In addition, various, different objects have to be picked up by means of the robot arm and transported to a different location.

 

The Humanoid League (TeenSize):

In the Humanoid League, on the other hand, it is about football. At the same time, robots that are built similar to humans are played together according to the rules of FIFA. The aim of the Humanoid League is to compete against the then reigning FIFA world champion of the people in a regular football match in 2050.
In the Humanoid League, three sizes of robots are distinguished: KidSize, TeenSize and AdultSize. The sizes range from very small to almost man-sized. The team of the Ostfalia has medium-sized robots and plays in the TeenSize.

 

Source: https://www.ostfalia.de/cms/de/campus/ostfalia/news/news0272.html

Humanoid – Bot Evolution

Posted on Posted in Humanoid-en

detlef

 Detlef

since June 2014

Rule changes for 2014 have caused us to build larger robots. For this, we have adapted the Nimbro OP to a size of 85 cm. The aluminum and carbon parts were manufactured as before in the university workshop. We created plastic parts, like the head and the hands, with our 3D printer. In addition to significantly larger batteries are used Intel NUCs with i5 processor.

 

 

 

 

da-v1n

Da-v1n

since June 2012

The mechanical design of this generation is based on the DARwIn-OP. With a sturdy aluminum scaffolding and stronger servos (RX-28 or MX-28), we have a convincing new platform. We have the metal parts made in our university workshop. On a new main board is now an Intel Atom processor is used.

 

 

 

 

copper-2011_evo

Copper

September 2010 – June 2012

In addition to some small improvements, two new degrees of freedom have been added.

 

 

 

 

 

johnny-neu

 Johnny_v2

June 2009 – September 2010

After we noticed on the GermanOpen 2009 that the robot team around JonnyFirst is too heavy, we have dismantled the PDA and now work in the head with an embedded Linux. Both image processing and artificial intelligence are now running on the processor. With this model, we entered the RoboCup 2009 in Graz.

 

 

 

 

 

german-open-johnny

Johnny_v1 

  April 2009 – June 2009

With the improved version of Jonny we participated in the GermanOpen 2009. Here we have reduced the head and also made the legs smaller again. Much has also been done on electronics. So we had only 3 controllers at this time.

 

 

 

 

 

huno

 Huno 

February 2009 – April 2009

With the so-called “Huno”, the humanoid construction from the RoboBuilder Kit, we have continued. It was equipped with image processing and PDA and we shot our first goal. With JonnyFirst, our extended version of the RoboBuilder kit, we have shot our qualification video.

 

 

 

 

bioloid

 Bioloid 

End of 2008 – February 2009

We made our first attempts with humanoid robots with the Bioloid Kit. We have nevertheless decided, for reasons of cost and since the wCK servos faster, for the RoboBuilder kit.

RoboCup China 2015

Posted on Posted in News-en
 Humanoid League

Everything started with our luggage hanging in Guangzhou. So we did not just lack robots and equipment, but above all clothes for changing after the long journey. What remained to us was the hand luggage, with which we had at least one or the other, if of course no tool. In the hotel we met our partners from Brazil, who had arrived with two robots. On the first day of preparation we still missed two of the big robots, which we could not prepare. The first one, which had arrived, we did not get through the robot check until the head was too big. It was measured differently than in the previous year, which caused problems not only with us. So we had to cut our head, which looked ugly and also led to more instability. Even the Brazilian comrade had to take off his helmet and was not allowed to play with it. While the Brazilians were working on the new running algorithm, our people were in the process of calibrating the recognition of objects. But the time that remained was short and was not enough. While the recognition of the playing field and the goal post worked well, especially the search for the ball of the Teen Size brought unimagined problems. This involved neglecting running and shooting, which led to frequent falls in the games.

 

In the Kid Size, we were in group B with ZJUDancer and FUmanoids, which did not allow us an easy start. The final result of the group phase was then no surprise. In the intermediate round we had once again the chance against RoboErectus, where our robots unfortunately could not get through. In Teen Size, we had similar difficulties and could not score goals. Overall the competition was stressful for us because we had one game after another by participating in both leagues. It was noticeable that we were short of specialists and robots. In the Technical Challenges, however, we were once again able to prove what our robots are capable of doing. They could not only jump surprisingly high, but also balance large shocks of the push recovery. We were able to secure second place in both leagues.

Humanoid League

Posted on Posted in Humanoid-en

da-v4nSince 2009 we are represented in the Humanoid Kid Size League. The special thing about this league is that the robots have to be very similar to humans. For example, they must run on two legs and have physical-human characteristics. As a further difficulty, only sensors are allowed, which also man has. We use cameras (eyes), gyroscope and accelerometer (balance sense) and some of the other robots also have pressure sensors on the feet.

Other special sensors are not permitted. Furthermore, the robots must also play completely autonomously; Which means that the intelligence must also be in the robot and not on another computer, which in turn turns the robot remotely. The robots are allowed to talk to each other via WLAN to discuss themselves.

In the Humanoid League there are two areas: the Games and the Technical Challenges. In the games, four are played against four football, the rules are modeled on the FIFA rules and are only slightly simplified, so that a game can come about. For this, the robots have to bring a whole lot of abilities to play at all: they must be able to move and shoot, stand up independently and of course find the ball.

Game

game-smallThe Technical Challenges is about testing and working on things that are not yet in the game. In addition, techniques that are currently available in human football are to be learned and improved. Currently, there are four Technical Challenges. Obstacle Avoidance and Dribbling is about dribbling the ball through obstacles in the goal. In doing so, the abilities are identified as obstacles and a path through the obstacles is to be found and planed.

When throwing in, the ability to lift a ball and throw it is thrown. The goal is to integrate the throw into the game as soon as possible, in order to get closer to human football. The double pass is used to promote the abilities “obstacles”, “recognition of fellow players” and “closer shooting”. Another challenge is the high-kick, where the robot is to reach the goal as high as possible.

German Open 2015

Posted on Posted in News-en

Humanoid League

This year’s German Open was a little spectacular, but with many challenges on the ground. By the new rules – artificial grass, a new white-colored ball and white gates – the robots demanded everything. We were with two small and two large robots at the start, could because of unexpected problems with the hardware but unfortunately only partially one or two at the same time. The initial round robin phase went all dead, finding the ball and then moving to the ball was difficult not only for us at first. Then we came to the Penalty Shootout, where we secured fourth place. It was followed by the semi-final against the Fumanoids, which went 1: 0 for them, so we came into the game for the third place. In this game, we were able to show that we can not only run stably, but are even capable of shooting shots without adding. Nevertheless, all the arts were not enough for a goal during the game, it went 0-0. In the following Penalty Shootout our robots had the problem of recognizing the ball and running past the ball again and again. In the last attempt, our opponents, Bold Hearts, were able to move the ball a little way forward while we have not reached the ball. This was ultimately the decision and the fourth place for us. The German Open has revealed some weaknesses and allows us to prepare ourselves for the World Cup in China.

RoboCup Brasili 2014

Posted on Posted in News-en

(By Prof. R. Gerndt)

The Eastfalia was represented by a team of 10 students who were reinforced by 4 students from our partner school in Santa Maria, Brazil, the Universidade Federal Santa Maria (UFSM).

In the ‘Humanoid League’, the Ostfalia after the preliminary round (1. ‘Round Robin’) by a single own goal in the first knockout round. Apart from general problems in the preparation, the delayed delivery of a suitcase with numerous tools also caused problems. The goal to get into the second round (2nd ‘Round Robin’) was missed. The team has, however, for the first time participated in some ‘Technical Challenges’ and was able to score points. On the RoboCup2014, a new, larger robot and a robot equipped with elastic knee drives (Series Elastic Actuator – SEA) were also used for the first time.

The RoboCup @ work team had to deal with hardware and software problems. Already at the beginning of the competitions, for example, the gripper had failed and had to be replaced by a loan from another team. In addition, the “Conveyor belt challenge” had to be omitted for technical reasons, which brought additional disadvantages to the team, who could last this test as the only team. The goal of 1000 points in the competitions (approx. Third place) was not reached with about 680 points.

At the RoboCup Symposium (http://www.robocup2014.org/?page_id=27) Ostfalia was represented with a research project on “Series Elastic Actuator”, which was jointly funded by the UFSM. The importance of the topic was underlined by the two invited lectures by Prof. Brooks (MIT) and Prof. Haddadin (Leibniz University). The response to the Ostfalia contribution was correspondingly large. The topic is to be pursued intensively.

In the run-up to the RoboCup is a contribution for the NDR television (http://www.ndr.de/fernsehen/sendungen/hallo_niedersachsen/Fussball-Roboter-wollen-Weltmeistertitel-holen,hallonds22546.html). The SEA robot was mentioned in the heise-newsticker (http://www.heise.de/newsticker/meldung/RoboCup-WM-2016-kicken-die-Roboter-in-Leipzig-2264401.html?hg=1&hgi=4&hgf = False).

The contribution of the Ostfalia to the further development of the RoboCup was recognized. Mr. Hitzmann was confirmed as a member of the Organizing Committee (OC) for RoboCup2015. Prof. Gerndt has been appointed to the executive committee (‘Exec’) for the ‘Humanoid League’ (for the structure see http://www.robocup.org/organization-of-robocup/).

In all, a number of contacts were made. The team has received invitations to study stays and cooperation with other universities. The SEA research project is to be expanded further in a close cooperation between Ostfalia and UFSM.

23.07.2014

Unfortunately, our humanoid robots were eliminated yesterday by an own goal from Arctos with 0: 1.

After the last game, Lupo’s leg had to be repaired. Unfortunately, due to other problems we had too little time to test, where he could still run and shoot. Unfortunately, his shot from the left flank was a bit of a pain, but he stayed up unsuccessfully without a problem.

After the game, the team needed some encouragement before it started the preparations for the Technical Challenges. This is why we have created the emergence of our large humanoid robot. After he got up like our little humanoids, we went on with the walk (Oliver had already programmed it). In the test on the normal playing field the walk was already very good, so we went directly to the field with artificial grass. There, too, the robot could run to our delight.

20.07.2014

On Friday, we arrived at João Pessoa after three flights and welcomed the Brazilian students Leandro, Milton and Fernando into our group.

Unfortunately, one of our suitcases has been missing since the last flight. Since it is the humanoid toolkit (including cold sprays, multiple sockets and keyboards), Benny and Milton had to try to buy some work tools, and the teams were able to use the existing tools. This worked well, so that the loss of the cold spray is the biggest loss since we can not buy this on the spot.

Despite these slight difficulties, the humanoids were able to successfully use the first setup day and the four small humanoid robots were almost finished in the evening. In our apartment were still a few popped errors of the software, which was played this morning on the robots.

 

As can be seen in the picture above, the humanoids have been able to continue working on their large robot. This morning he was allowed to play for the first time in Brazil on one of the football fields.

With our @ WORK team ran yesterday the arm not, but after a day of hard work and with the help of an older version, we have not only fixed this, but started to optimize a lot. This includes, for example, an acceleration of the arm’s alignment with objects. Our goal is to arm the YouBots arm more than twice as fast!

Educational prize alliance for the region

Posted on Posted in News-en

On 04.12.2014 in Wolfenbüttel the education prize was awarded by Allianz for the region. This award is an award given by a jury of educational institutions and business enterprises. Our team participated in this competition and successfully competed against the other 119 projects and took 3rd place in the category “Adults”. The award takes place at regular intervals and praises a total sum of 30,000 €.

In the sense, the team says thank you!

New Humanoid Robots from Eckensberger Foundation

Posted on Posted in News-en

Today we have received information from the Eckensberger Foundation: They sponsor us in the creation of two new humanoid robots. We are very pleased with this support and have immediately ordered the components. The new humanoid robots will be bigger than our current ones and we hope that we can take one of the two new robots to the World Cup to Brazil.