Learning symmetric and low-energy locomotion
Explore this paper's citation graph
Summary
A minimalist learning approach to the locomotion problem is taken, without the use of motion examples, finite state machines, or morphology-specific knowledge, to produce locomotion behaviors that are symmetric, low-energy, and much closer to that of a real person.
- Type
- article
- Published
- 2018-01-24
- Cited by
- 229
- References
- 66
- Access
- Open access
- OpenAlex
- https://openalex.org/W2784449021
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:21747305
Keywords
Benchmark (surveying), Reinforcement learning, Character (mathematics), Motion (physics), Function (biology)
References
- Stable Proportional-Derivative Controllers
- Curriculum learning of multiple tasks
- Data-driven biped control
- Feature-based locomotion controllers
- Asymmetries in ground reaction force patterns in normal human gait.
- Continuation methods for adapting simulated skills
- Iterative Training of Dynamic Skills Inspired by Human Coaching Techniques
- SIMBICON: simple biped locomotion control
- Locomotion control for many-muscle humanoids
- Learning bicycle stunts
- Optimization-based interactive motion synthesis
- Dynamic terrain traversal skills using reinforcement learning
- Generalized biped walking control
- Animating human lower limbs using contact-invariant optimization
- Interactive Character Animation Using Simulated Physics: A State‐of‐the‐Art Review
- Simulating biped behaviors from human motion data
- Gait asymmetry in community-ambulating stroke survivors.
- Discovery of complex behaviors through contact-invariant optimization
- Locomotion skills for simulated quadrupeds
- Optimal feedback control for character animation using an abstract model
Cited by
- Feedback Control For Cassie With Deep Reinforcement Learning
- SFV
- Generating Animated Videos of Human Activities from Natural Language Descriptions
- Interactive character animation by learning multi-objective control
- Learning to dress
- Aerobatics control of flying creatures via self-regulated learning
- Learning agile and dynamic motor skills for legged robots
- Learning Whole-Body Motor Skills for Humanoids
- Sim-to-Real Transfer for Biped Locomotion
- Iterative Reinforcement Learning Based Design of Dynamic Locomotion Skills for Cassie
- Synthesis of biologically realistic human motion using joint torque actuation
- Data Efficient and Safe Learning for Locomotion via Simplified Model
- Scalable muscle-actuated human simulation and control
- Towards Robust Direction Invariance in Character Animation
- Energy-Efficient Slithering Gait Exploration for a Snake-like Robot based on Reinforcement Learning
- Self-Imitation Learning of Locomotion Movements through Termination Curriculum
- Convolutional Humanoid Animation via Deformation
- Deep Learned Path Planning via Randomized Reward-Linked-Goals and Potential Space Applications
- Learning a Control Policy for Fall Prevention on an Assistive Walking Device
- On Learning Symmetric Locomotion
Related papers
No related papers recorded.