Parkour Learning for Quadrupeds via Terrain-Conditional Adversarial Motion Priors

Autor

Artículo de revista

Idioma

Agile parkour in unstructured environments poses significant challenges for quadruped robots, requiring both dynamic motion generation and terrain adaptability. Recent advances such as Adversarial Motion Priors (AMP) have shown promise in learning dynamic behaviors through motion imitation, but the resulting policies are typically specialized and struggle to generalize across varying terrains. However, existing AMP-based approaches largely lack explicit environmental awareness, leading to limited adaptability and revealing a clear gap in achieving general agile locomotion. To address this limitation, we propose a novel terrain-conditional AMP framework that extends adversarial motion priors by conditioning the discriminator on explicit terrain features, enabling the learning of terrain-aware motion representations adaptable to diverse environments. To improve practical applicability, we further leverage a vision-based policy distillation scheme, where a teacher policy with privileged terrain height information supervises a student policy operating only on forward-looking depth images. This enables agile, perception-driven locomotion in real time. To the best of our knowledge, this is the first work to integrate environmental information into adversarial motion priors and jointly learn a vision-based policy through policy distillation for agile quadruped locomotion. Experiments on terrains such as platforms, gaps, stairs, slopes, and debris show that the proposed method achieves more efficient training convergence and higher success rates compared to pure AMP-based and RL-based methods. These results highlight the effectiveness of the proposed framework and represent a step toward perception-driven agile locomotion for quadruped robots in complex environments.

URL

PUMA: Perception-Driven Unified Foothold Prior for Mobility Augmented Quadruped Parkour

Autor

Artículo de revista

Idioma

Parkour tasks for quadrupeds have emerged as a promising benchmark for agile locomotion. While human athletes can effectively perceive environmental characteristics to select appropriate footholds for obstacle traversal, endowing legged robots with similar perceptual reasoning remains a significant challenge. Existing methods often rely on hierarchical controllers that follow pre-computed footholds, thereby constraining the robot’s real-time adaptability and the exploratory potential of reinforcement learning. To overcome these challenges, we present PUMA, an end-to-end learning framework that integrates visual perception and foothold priors into a single-stage training process. This approach leverages terrain features to estimate egocentric polar foothold priors, composed of relative distance and heading, guiding the robot in active posture adaptation for parkour tasks. Extensive experiments conducted in simulation and real-world environments across various discrete complex terrains demonstrate PUMA’s exceptional agility and robustness in challenging scenarios.

URL

Exploring and Evaluating Prompt Engineering Techniques for Game Development: Using a Rhythm-Based Parkour Game as a Test Case

Autor

Publicación en congreso

Idioma

This study explores and evaluates prompt engineering techniques in game development using Large Language Models (LLMs). As LLMs become increasingly prevalent in technical domains, constructing effective prompts is crucial for generating high-quality game code. Through a series of progressively complex prompts, this study assessed how different prompt structures influence the quality of games generated by Claude and ChatGPT. The findings demonstrate a strong positive correlation between prompt complexity and game quality, with Claude showing significantly higher responsiveness to detailed specifications compared to ChatGPT. A comprehensive evaluation framework is designed using five criteria: Start Interface, Game Interface, Playability, Rhythm Integration, and Game Completeness, providing a standardized method for assessing AI-generated games. The research reveals that prompts incorporating clear technical requirements, gameplay mechanics specifications, and user experience considerations yield optimal results. However, diminishing returns were observed at extreme levels of prompt complexity, suggesting an ideal balance between detail and conciseness. These insights contribute to the understanding of effective LLM utilization in creative technical fields and establish a foundation for future research exploring prompt engineering across various game genres and frameworks as LLM capabilities continue to advance.

URL

Acrobotics: A Generalist Approach to Quadrupedal Robots' Parkour

Autor

Publicación en congreso

Idioma

Climbing, crouching, bridging gaps, and walking up stairs are just a few of the advantages that quadruped robots have over wheeled robots, making them more suitable for navigating rough and unstructured terrain. However, executing such manoeuvres requires precise temporal coordination and complex agent-environment interactions. Moreover, legged locomotion is inherently more prone to slippage and tripping, and the classical approach of modeling such cases to design a robust controller thus quickly becomes impractical. In contrast, reinforcement learning offers a compelling solution by enabling optimal control through trial and error. We present a generalist reinforcement learning algorithm for quadrupedal agents in dynamic motion scenarios. The learned policy rivals state-of-the-art specialist policies trained using a mixture of experts approach, while using only 25% as many agents during training. Our experiments also highlight the key components of the generalist locomotion policy and the primary factors contributing to its success.

URL

Symmetric Parkour Learning for Robust Quadruped Robot Control on Challenging Terrain

Autor

Artículo de revista

Idioma

​

URL

Model Predictive Parkour Control of a Monoped Hopper in Dynamically Changing Environments

Autor

Artículo de revista

Idioma

A great advantage of legged robots is their ability to operate on particularly difficult and obstructed terrain, which demands dynamic, robust, and precise movements. The study of obstacle courses provides invaluable insights into the challenges legged robots face, offering a controlled environment to assess and enhance their capabilities. Traversing it with a one-legged hopper introduces intricate challenges, such as planning over contacts and dealing with flight phases, which necessitates a sophisticated controller. A novel model predictive parkour controller is introduced, that finds an optimal path through a real-Time changing obstacle course with mixed integer motion planning. The execution of this optimized path is then achieved through a state machine employing a PD control scheme with feedforward torques, ensuring robust and accurate performance.

URL

Learning Visual Parkour from Generated Images

Autor

Publicación en congreso

Idioma

Fast and accurate physics simulation is an essential component of robot learning, where robots can explore failure scenarios that are difficult to produce in the real world and learn from unlimited on-policy data. Yet, it remains challenging to incorporate RGB-color perception into the sim-to-real pipeline that matches the real world in its richness and realism. In this work, we train a robot dog in simulation for visual parkour. We propose a way to use generative models to synthesize diverse and physically accurate image sequences of the scene from the robot's ego-centric perspective. We present demonstrations of zero-shot transfer to the RGB-only observations of the real world on a robot equipped with a low-cost, off-the-shelf color camera.

URL

Extreme Parkour with Legged Robots

Autor

Artículo de revista

Idioma

Humans can perform parkour by traversing
obstacles in a highly dynamic fashion requiring precise eyemuscle coordination and movement. Getting robots to do the
same task requires overcoming similar challenges. Classically,
this is done by independently engineering perception, actuation,
and control systems to very low tolerances. This restricts
them to tightly controlled settings such as a predetermined
obstacle course in labs. In contrast, humans are able to learn
parkour through practice without significantly changing their
underlying biology. In this paper, we take a similar approach
to developing robot parkour on a small low-cost robot with
imprecise actuation and a single front-facing depth camera
for perception which is low-frequency, jittery, and prone to
artifacts. We show how a single neural net policy operating
directly from a camera image, trained in simulation with largescale RL, can overcome imprecise sensing and actuation to
output highly precise control behavior end-to-end. We show
our robot can perform a high jump on obstacles 2x its height,
long jump across gaps 2x its length, do a handstand and run
across tilted ramps, and generalize to novel obstacle courses
with different physical properties

URL

Parkour,movimiento y ciudad. Cuerpo humano en movimiento en el entorno construido

Autor

Trabajo Fin de Grado/Máster o similar

Idioma

El siguiente trabajo desarrolla un estudio cualitativo de los espacios que se utilizan para practicar parkour, para ello se sirve de una serie de casos de estudio que recogen desde espacios urbanos utilizados recurrentemente por los practicantes de parkour hasta instalaciones que han sido expresamente construidas para practicar esta disciplina, todos ellos ubicados en la ciudad de Madrid. Se han llevado a cabo levantamientos de cada caso de estudio y se han analizado teniendo en cuenta tanto sus variables físicas como el modo en que son empleados, detectando tanto sus fortalezas como sus debilidades en relación a la práctica del parkour. Con la información obtenida de los casos de estudio mediante un análisis comparativo, unida a la confección de una figura humana de referencia basada en datos antropométricos contrastados y la información extraída de la encuesta incluida en este trabajo dirigida a traceurs y traceurses, se elabora una guía de diseño que tiene por objeto exponer todas las variables implícitas en el diseño de un espacio destinado a la práctica del parkour, desde su implantación en la ciudad hasta su relación con el cuerpo humano, apoyándose en dibujos que explican el dimensionado de los obstáculos y tablas de medidas donde se recogen las distancias que pueden cubrirse en función del tipo de salto.

URL

Analysis of Fluency of Movement in Parkour Using a Video and Inertial Measurement Unit Technology

Autor

Artículo de revista

Idioma

Fluency is a movement parameter combining smoothness and hesitation, and its objective measurement may be used to determine the effects of practice on sports performance. This study aimed to measure fluency in parkour, an acrobatic discipline comprising complex non-cyclical movements, which involves fluency as a critical aspect of performance. Inter-individual fluidity differences between advanced and novice athletes as well as intra-individual variations of fluency between different parts and subsequent repetitions of a path were addressed. Seventeen parkour participants were enrolled and divided into two groups based on their experience. We analysed signals captured from an inertial measurement unit fixed on the back of the pelvis of each participant during three consecutive repetitions of a specifically designed parkour routine under the guidance of video analysis. Two fluency parameters, namely smoothness and hesitation, were measured. Smoothness was calculated as the number of inflexions on the so-called jerk graph; hesitation was the percentage of the drop in the centre of mass velocity. Smoothness resulted in significantly lower values in advanced athletes (mean: 126.4; range: 36-192) than in beginners (mean: 179.37; range: 98-272) during one of the three motor activities (p = 0.02). A qualitative analysis of hesitation showed that beginner athletes tended to experience more prominent velocity drops and negative deflection than more advanced athletes. In conclusion, a system based on a video and an inertial measurement unit is a promising approach for quantification and the assessment of variability of fluency, and it is potentially beneficial to guide and evaluate the training process.

URL

Acrobotics: A Generalist Approach to Quadrupedal Robots' Parkour

Autor

Publicación en congreso

Idioma

Climbing, crouching, bridging gaps, and walking up stairs are just a few of the advantages that quadruped robots have over wheeled robots, making them more suitable for navigating rough and unstructured terrain. However, executing such manoeuvres requires precise temporal coordination and complex agent-environment interactions. Moreover, legged locomotion is inherently more prone to slippage and tripping, and the classical approach of modeling such cases to design a robust controller thus quickly becomes impractical. In contrast, reinforcement learning offers a compelling solution by enabling optimal control through trial and error. We present generalist reinforcement learning algorithm for quadrupedal agents in dynamic motion scenarios. The learned policy rivals state-of-the-art specialist policies trained using a mixture of experts approach, while using only 25% as many agents during training. Our experiments also highlight the key components of the generalist locomotion policy and the primary factors contributing to its success.

URL

Parkour in the wild: Learning a general and extensible agile locomotion policy using multi-expert distillation and RL fine-tuning

Autor

Artículo de revista

Idioma

Legged robots are well-suited for navigating terrains inaccessible to wheeled robots, making them ideal for applications in search and rescue or space exploration. However, current control methods often struggle to generalize across diverse, unstructured environments. This paper introduces a novel framework for agile locomotion of legged robots by combining multi-expert distillation with reinforcement learning (RL) fine-tuning to achieve robust generalization. Initially, terrain-specific expert policies are trained to develop specialized locomotion skills. These policies are then distilled into a unified generalist policy via the DAgger algorithm. The distilled policy is subsequently fine-tuned using RL on a broader terrain set, including real-world 3D scans. The framework allows further adaptation to new terrains through repeated fine-tuning. The proposed policy leverages depth images as exteroceptive inputs, enabling robust navigation across diverse, unstructured terrains. Experimental results demonstrate significant performance improvements over existing methods in synthesizing multi-terrain skills into a single controller. Deployment on the ANYmal D robot validates the policy’s ability to navigate complex environments with agility and robustness, setting a new benchmark for legged robot locomotion.

URL

ES-Parkour: Advanced Robot Parkour with Bio-inspired Event Camera and Spiking Neural Network

Autor

Publicación en congreso

Idioma

In recent years, quadruped robotics has advanced significantly, particularly in perception and motion control via reinforcement learning, enabling complex motions in challenging environments. Visual sensors like depth cameras enhance stability and robustness but face limitations, such as low operating frequencies relative to joint control and sensitivity to lighting, which hinder outdoor deployment. Additionally, deep neural networks in sensor and control systems increase computational demands. To address these issues, we introduce spiking neural networks (SNNs) and event cameras to perform a challenging quadruped parkour task. Event cameras capture dynamic visual data, while SNNs efficiently process spike sequences, mimicking biological perception. Experimental results demonstrate that this approach significantly outperforms traditional models, achieving excellent parkour performance with just 11.7% of the energy consumption of an artificial neural network (ANN)-based model, yielding an 88.3% energy reduction. By integrating event cameras with SNNs, our work advances robotic reinforcement learning and opens new possibilities for applications in demanding environments.

URL

Microrobot Vascular Parkour: Analytic Geometry-based Path Planning with Real-time Dynamic Obstacle Avoidance

Autor

Publicación en congreso

Idioma

Autonomous microrobots in blood vessels could enable minimally invasive therapies, but navigation is challenged by dense, moving obstacles. We propose a real-time path planning framework that couples an analytic geometry global planner (AGP) with two reactive local escape controllers, one based on rules and one based on reinforcement learning, to handle sudden moving obstacles. Using real-time imaging, the system estimates the positions of the microrobot, obstacles, and targets and computes collision-free motions. In simulation, AGP yields shorter paths and faster planning than weighted A* (WA*), particle swarm optimization (PSO), and rapidly exploring random trees (RRT), while maintaining feasibility and determinism. We extend AGP from 2D to 3D without loss of speed. In both simulations and experiments, the combined global planner and local controllers reliably avoid moving obstacles and reach targets. The average planning time is 40 ms per frame, compatible with 25 fps image acquisition and real-time closed-loop control. These results advance autonomous microrobot navigation and targeted drug delivery in vascular environments.

URL

ANYmal parkour: Learning agile navigation for quadrupedal robots

Autor

Artículo de revista

Idioma

Performing agile navigation with four-legged robots is a challenging task because of the highly dynamic motions, contacts with various parts of the robot, and the limited field of view of the perception sensors. Here, we propose a fully learned approach to training such robots and conquer scenarios that are reminiscent of parkour challenges. The method involves training advanced locomotion skills for several types of obstacles, such as walking, jumping, climbing, and crouching, and then using a high-level policy to select and control those skills across the terrain. Thanks to our hierarchical formulation, the navigation policy is aware of the capabilities of each skill, and it will adapt its behavior depending on the scenario at hand. In addition, a perception module was trained to reconstruct obstacles from highly occluded and noisy sensory data and endows the pipeline with scene understanding. Compared with previous attempts, our method can plan a path for challenging scenarios without expert demonstration, offline computation, a priori knowledge of the environment, or taking contacts explicitly into account. Although these modules were trained from simulated data only, our real-world experiments demonstrate successful transfer on hardware, where the robot navigated and crossed consecutive challenging obstacles with speeds of up to 2 meters per second.

URL

Humanoid Parkour Learning

Autor

Artículo de revista

Idioma

Parkour is a grand challenge for legged locomotion, even for quadruped robots, requiring active perception and various maneuvers to overcome multiple challenging obstacles. Existing methods for humanoid locomotion either optimize a trajectory for a single parkour track or train a reinforcement learning policy only to walk with a significant amount of motion references. In this work, we propose a framework for learning an end-to-end vision-based whole-body-control parkour policy for humanoid robots that overcomes multiple parkour skills without any motion prior. Using the parkour policy, the humanoid robot can jump on a 0.42m platform, leap over hurdles, 0.8m gaps, and much more. It can also run at 1.8m/s in the wild and walk robustly on different terrains. We test our policy in indoor and outdoor environments to demonstrate that it can autonomously select parkour skills while following the rotation command of the joystick. We override the arm actions and show that this framework can easily transfer to humanoid mobile manipulation tasks

URL

SoloParkour: Constrained Reinforcement Learning for Visual Locomotion from Privileged Experience

Autor

Publicación en congreso

Idioma

Parkour poses a significant challenge for legged robots, requiring navigation through complex environments with agility and precision based on limited sensory inputs. In this work, we introduce a novel method for training end-to-end visual policies, from depth pixels to robot control commands, to achieve agile and safe quadruped locomotion. We formulate robot parkour as a constrained reinforcement learning (RL) problem designed to maximize the emergence of agile skills within the robot's physical limits while ensuring safety. We first train a policy without vision using privileged information about the robot's surroundings. We then generate experience from this privileged policy to warm-start a sample efficient off-policy RL algorithm from depth images. This allows the robot to adapt behaviors from this privileged experience to visual locomotion while circumventing the high computational costs of RL directly from pixels. We demonstrate the effectiveness of our method on a real Solo-12 robot, showcasing its capability to perform a variety of parkour skills such as walking, climbing, leaping, and crawling.

URL

PIE: Parkour With Implicit-Explicit Learning Framework for Legged Robots

Autor

Artículo de revista

Idioma

Parkour presents a highly challenging task for legged robots, requiring them to traverse various terrains with agile and smooth locomotion. This necessitates comprehensive understanding of both the robot's own state and the surrounding terrain, despite the inherent unreliability of robot perception and actuation. Current state-of-the-art methods either rely on complex pre-trained high-level terrain reconstruction modules or limit the maximum potential of robot parkour to avoid failure due to inaccurate perception. In this paper, we propose a one-stage end-to-end learning-based parkour framework: Parkour with Implicit-Explicit learning framework for legged robots (PIE) that leverages dual-level implicit-explicit estimation. With this mechanism, even a low-cost quadruped robot equipped with an unreliable egocentric depth camera can achieve exceptional performance on challenging parkour terrains using a relatively simple training process and reward function. While the training process is conducted entirely in simulation, our real-world validation demonstrates successful zero-shot deployment of our framework, showcasing superior parkour performance on harsh terrains.

URL

Robot Parkour Learning

Autor

Publicación en congreso

Idioma

Parkour is a grand challenge for legged locomotion that requires robots to overcome various obstacles rapidly in complex environments. Existing methods can generate either diverse but blind locomotion skills or vision-based but specialized skills by using reference animal data or complex rewards. However, autonomous parkour requires robots to learn generalizable skills that are both vision-based and diverse to perceive and react to various scenarios. In this work, we propose a system for learning a single end-to-end vision-based parkour policy of diverse parkour skills using a simple reward without any reference motion data. We develop a reinforcement learning method inspired by direct collocation to generate parkour skills, including climbing over high obstacles, leaping over large gaps, crawling beneath low barriers, squeezing through thin slits, and running. We distill these skills into a single vision-based parkour policy and transfer it to a quadrupedal robot using its egocentric depth camera. We demonstrate that our system can empower two different low-cost robots to autonomously select and execute appropriate parkour skills to traverse challenging real-world environments

URL

Extreme Parkour with Legged Robots

Autor

Artículo de revista

Idioma

Humans can perform parkour by traversing obstacles in a highly dynamic fashion requiring precise eye-muscle coordination and movement. Getting robots to do the same task requires overcoming similar challenges. Classically, this is done by independently engineering perception, actuation, and control systems to very low tolerances. This restricts them to tightly controlled settings such as a predetermined obstacle course in labs. In contrast, humans are able to learn parkour through practice without significantly changing their underlying biology. In this paper, we take a similar approach to developing robot parkour on a small low-cost robot with imprecise actuation and a single front-facing depth camera for perception which is low-frequency, jittery, and prone to artifacts. We show how a single neural net policy operating directly from a camera image, trained in simulation with largescale RL, can overcome imprecise sensing and actuation to output highly precise control behavior end-to-end. We show our robot can perform a high jump on obstacles 2x its height, long jump across gaps 2x its length, do a handstand and run across tilted ramps, and generalize to novel obstacle courses with different physical properties.

URL