Project Description

Robust legged platform for highly dynamic motions

Kangaroo robot is a prototype bipedal robot research platform for exploring advanced control methods for legged locomotion. The robot is lightweight with low moving inertia in the legs and suitable for highly dynamic motions such as jumping and running. Each leg has 6 DoF, and a mass of 12.5 kg mostly located in the upper part so that the mechanical design allows for impact resilience and reduced energy consumption.

Side view of the robot Kangaroo
Frontal view of the biped robot Kangaroo

Linear actuators and non-linear transmissions

Kangaroo incorporates custom linear actuators with integrated force sensors providing a wide range of motion as well as high speed and torques at the joints. The design takes advantage of nonlinear linkage mechanisms to place all the actuators close to the trunk, making the biped robot research platform closer to classical template models. The non-linear transmission parameters have been optimised to achieve the necessary speeds and torques for jumping but require only low power whilst standing and walking.

Leg length actuator designed for efficiency

Kangaroo robot uses an actuation system for the extension and contraction of the leg that is implemented with only one motor, in contrast to other robots that use multiple actuators at the hip, knee, and ankle. The decoupling of the leg length and hip swing motions with Kangaroo makes the system highly efficient.

Feet and legs of the research platform Kangaroo
Motor of Kangaroo robot

Advanced motor control technology

Kangaroo uses a new generation of in-house electronics boards for motor control and sensor acquisition. This choice allows embedding custom and decentralized closed loop force/torque controllers at joint level and provides flexibility for experimenting with different control algorithms and obtaining the best performance from the biped robot research platform.

Areas of use

USED FOR RESEARCH IN:

  • Artificial Intelligence

  • Machine learning

  • Control algorithms

  • Navigation
  • Torque Control

  • Highly reactive and dynamic motion
  • Dynamic Walking
  • Trajectory Optimization
  • High impact resilience

  • Motion planning

FIELDS OF APPLICATION:

Factory of the Future

IoT

Benchmarking

Space exploration

space_exploration

Contact PAL Robotics team if you have any further questions!