LOLA

What it's for

LOLA, a 176 cm bipedal platform from the Chair of Applied Mechanics at the Technical University of Munich (TUM), was built to answer one question: how do you teach a machine to walk like a human, fast and fluidly, even on uneven terrain?

Research on fast walking. The design has 26 degrees of freedom (DOF) and legs with redundant, seven-DOF kinematics. The redundancy gives extra room to move: the robot can vary the positions of its knee, hip and ankle to avoid an obstacle or catch its balance. Researchers who plan motion in real time and test control algorithms need exactly that. For the basics of how a humanoid walks at all, see our article “How Does a Humanoid Robot Work”.

Walking on uneven ground. This is the hardest area and, according to the project description, LOLA's main strength: stability on difficult terrain. In theory, this kind of research could be useful for rescue and construction robots, but we found no information on LOLA being used outside the lab. For comparison, look at the quadruped robots in “Quadruped Robots: Uses in Inspection, Security and Industry”, which handle uneven terrain commercially today.

Motion planning and control. The university uses the robot for work on motion dynamics; the results of such work usually end up in scientific publications. For background, see “Actuators in Humanoid Robots”.

Construction: brushless electric motors in a modular, lightweight joint layout drive the robot, and the total mass is 68 kg. Sensors and the onboard computer are not described.

This is not a robot for shop-floor work. It is a scientific platform worth knowing because it shows where commercial bipeds such as Unitree H1 or Unitree H1-2 come from. Not available in Poland; application areas: universities and laboratories.

Who it's for

Worth considering for: researchers in bipedal walking, PhD students and lecturers in mechanics and control; editorial teams that want to get to know the key academic projects in Europe; R&D companies looking for a benchmark for their own locomotion research.

Probably not for: anyone who wants to buy a robot. LOLA is not a product; it has no price, no distribution and no path to commercialization. Not for schools or hobbyists. For companies looking for a humanoid they can order, we recommend “Unitree H1, G1, R1 and H2”.

For mechanics students, however, it is an interesting topic for a review paper: LOLA shows how kinematic redundancy (seven degrees of freedom in the leg instead of the minimum six) translates into freedom in step planning. If you want to understand the terminology, start with “Degrees of Freedom (DOF) in Humanoid Robots” and “Robotics Glossary: 65 Terms with Polish Equivalents”.

Deployments

The robot is not a product and has no distributor.

Worldwide, LOLA operates in the TUM lab as a scientific platform; we found no information on commercial deployments or copies at other institutions. Polish universities usually have cheaper humanoids, which we cover in “Educational Robots for Schools and Universities”. For the domestic market, see “Polish Robotics Companies”.

WABIAN-2 is a Japanese research biped: 41 DOF, 148 cm, 64 kg, with a gait close to that of a human. Even more academic in character, with more degrees of freedom.

TORO is a German robot from DLR (since 2013): 174 cm, 76 kg, 39 DOF, research into dynamic walking. A very similar research mission.

Unitree H1 is a commercial humanoid you can actually buy; far easier to obtain than any university prototype, but without research documentation of this depth.

PAL Robotics TALOS is a research humanoid, 175 cm and 95 kg, with torque control in nearly every joint. It is sold, but runs only about 1.5 h.

Valkyrie is a NASA design from 2013: 187 cm, 136 kg, 44 degrees of freedom.

The Technical University of Munich, founded in 1868, develops LOLA at the Chair of Applied Mechanics. The launch date is not given, but the university's project page lists fast, autonomous, human-like walking and walking on uneven terrain as its goals.

We have no recorded events attached to LOLA, so we do not invent a timeline. What we can do is place it in the wider history: the makers of TORO (2013), WABIAN-2 and Valkyrie set themselves similar goals.