Marcell Missura | Rheinische Friedrich-Wilhelms-Universität Bonn (original) (raw)

Papers by Marcell Missura

Most of the research related to the topic of falling strategies considers falling to be an unavoi... more Most of the research related to the topic of falling strategies considers falling to be an unavoidable part of bipedal walking and is focused on developing strategies to avoid falls and to minimize mechanical damage. We take an alternative point of view and regard falling as a means to an end. We present our falling strategy for the specific case of a robot soccer goalie that deliberately jumps in front of a moving ball to prevent it from rolling into the goal. The jump decision is based on observed ball position, speed and direction of movement. We show how we implement a targeted falling into the appropriate direction, minimize the time from the jump decision to ground impact, and what solutions we developed to prevent mechanical damage. The presented falling technique was used in RoboCup Humanoid KidSize and TeenSize competitions and proved to be essential for winning. I. INTRODUCTION Falling is an inevitable part of bipedal walking, especially in dynamic environments such as rob...

The detection of planar surfaces in a point cloud is a popular technique for the extraction of dr... more The detection of planar surfaces in a point cloud is a popular technique for the extraction of drivable or walkable surfaces and for tabletop segmentation. Unfortunately, RGBD sensors are quite noisy and provide incomplete data, which makes the extraction of surfaces more challenging. Also, it is desirable to process the point cloud data in real time, which at a rate of approximately 30 Hz, leaves only a small amount of computation time per frame. We have already developed a real time-capable plane segmentation method [1] that exploits the organized structure of RGB-D point clouds in order to implement a computationally efficient region growing algorithm. It uses the point-plane distance to assign points to their segments rather than inherently unreliable surface normals. Now we are presenting an improvement where we adapt thresholds and other parameters of our algorithm to the measured depth in order to account for an increasing scatter of the points at larger distances from the ca...

Foresighted navigation is an essential skill for robots to rise from rigid factory floor installa... more Foresighted navigation is an essential skill for robots to rise from rigid factory floor installations to much more versatile mobile robots that partake in our every day environment. The current state of the art that provides this mobility to some extent is the Dynamic Window Approach combined with a global start-to-target path planner. However, neither the Dynamic Window Approach nor the path planner are equipped to predict the motion of other objects in the environment. We propose a change in the Dynamic Window Approach—a dynamic collision model—that is capable of predicting future collisions with the environment, even when objects are in motion. We show in our experiments that our new way of computing the Dynamic Window Approach excels in navigational fitness in dynamic environments while being computationally efficient.

2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017

We present a simple and efficient mathematical framework suitable for generating motion in the co... more We present a simple and efficient mathematical framework suitable for generating motion in the context of a variety of robotic motion tasks ranging from low-level motor control up to high-level locomotion planning. Our concept is based on a one-dimensional second-order model that allows analytic computation of its inverse dynamics while respecting physical constraints. This makes it a particularly useful tool for tasks that are expressed only as a start and goal state, such as animation key frames or way points in path planning. By means of time synchronization, the model extends easily to an arbitrary number of dimensions in a way that the target is reached in all dimensions at the same time. The framework excels in terms of execution time, which lies in the microsecond range even for high-dimensional trajectory generation tasks. We demonstrate our method in two different settings — full-body trajectory generation and path planning — and show its benefits in comparison with current...

Bipedal walking is one of the most essential skills in humanoid robot soccer. A stable and fast g... more Bipedal walking is one of the most essential skills in humanoid robot soccer. A stable and fast gait gives teams a winning edge when their robots are the first at the ball, maintain ball control with sure feet, and drive the ball decisively towards the opponent goal. The most successful teams in the Humanoid League are typically characterized by reliably walking robots. In this contribution, we describe the omnidirectional gait of team NimbRo, one of the most successful robot soccer teams in the history of RoboCup. The walk algorithm is open-loop and model-free. It is based on highly configurable, centralpattern-generated rhythmic motion signals and combines well with a compliant servo setting to achieve a relatively high level of self-stability. We discuss the advantages of this approach in comparison with methods of other successful teams and support our argumentation with experimental results.

ArXiv, 2018

Over the past few years, a number of successful humanoid platforms have been developed, including... more Over the past few years, a number of successful humanoid platforms have been developed, including the Nao and the DARwIn-OP, both of which are used by many research groups for the investigation of bipedal walking, full-body motions, and human-robot interaction. The NimbRo-OP is an open humanoid platform under development by team NimbRo of the University of Bonn. Significantly larger than the two aforementioned humanoids, this platform has the potential to interact with a more human-scale environment. This paper describes a software framework for the NimbRo-OP that is based on the Robot Operating System (ROS) middleware. The software provides functionality for hardware abstraction, visual perception, and behavior generation, and has been used to implement basic soccer skills. These were demonstrated at RoboCup 2013, as part of the winning team of the Humanoid League competition.

2021 IEEE International Conference on Robotics and Automation (ICRA)

The segmentation of a point cloud into planar primitives is a popular approach to first-line scen... more The segmentation of a point cloud into planar primitives is a popular approach to first-line scene interpretation and is particularly useful in mobile robotics for the extraction of drivable or walkable surfaces and for tabletop segmentation for manipulation purposes. Unfortunately, the planar segmentation task becomes particularly challenging when the point clouds are obtained from an inherently noisy, robot-mounted sensor that is often in motion, therefor requiring real time processing capabilities. We present a real time-capable plane segmentation technique based on a region growing algorithm that exploits the organized structure of point clouds obtained from RGB-D sensors. In order to counteract the sensor noise, we invest into careful selection of seeds that start the region growing and avoid the computation of surface normals whenever possible. We implemented our algorithm in C++ and thoroughly tested it in both simulated and real-world environments where we are able to compare our approach against existing state-of-the-art methods implemented in the Point Cloud Library. The experiments presented here suggest that our approach is accurate and fast, even in the presence of considerable sensor noise.

2021 IEEE International Conference on Robotics and Automation (ICRA)

Footstep planning is the dominating approach when it comes to controlling the walk of a humanoid ... more Footstep planning is the dominating approach when it comes to controlling the walk of a humanoid robot, even though a footstep plan is expensive to compute. The most prominent proposals typically spend up to a few seconds of computation time and output a sequence of up to 30 steps all the way to the goal. This way, footstep planning is applicable only in static environments where nothing changes after a plan has been computed. Since uncontrolled environments present challenges such as unforeseen motion of other objects and unexpected disturbances to balance, fast replanning of a footstep plan while the robot is in motion is highly desirable. We present a new way of fast footstep planning-Aborting A*-which is able to guarantee a replanning rate of 50 Hz by aborting an A* search before completion. We make aborting possible by using a novel, obstacle-aware heuristic function that lays out rotate-translaterotate motions along the shortest path to the goal, enabling us to stop the planning progress prematurely with a target-oriented solution at any time during the search, even after only a few nodes have been expanded. We show in our experiments that despite the bounded computation time, our planner computes good results and does not get stuck in local minima.

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

With the advent of polygonal maps finding their way into the navigational software of mobile robo... more With the advent of polygonal maps finding their way into the navigational software of mobile robots, the Visibility Graph can be used to search for the shortest collisionfree path. The nature of the Visibility Graph-based shortest path algorithms is such that first the entire graph is computed in a relatively time-consuming manner. Then, the graph can be searched efficiently any number of times for varying start and target state combinations with the A* or the Dijkstra algorithm. However, real-world environments are typically too dynamic for a map to remain valid for a long time. With the goal of obtaining the shortest path quickly in an ever changing environment, we introduce a rapid path finding algorithm-Minimal Construct-that discovers only a necessary portion of the Visibility Graph around the obstacles that actually get in the way. Collision tests are computed during an A* search only for lines that seem heuristically promising. This way, shortest paths can be found much faster than with a state-of-the-art Visibility Graph algorithm and as our experiments show, even grid-based A* searches are outperformed in most cases with the added benefit of smoother and shorter paths.

International Journal of Humanoid Robotics

Stable bipedal walking is a key prerequisite for humanoid robots to reach their potential of bein... more Stable bipedal walking is a key prerequisite for humanoid robots to reach their potential of being versatile helpers in our everyday environments. Bipedal walking is, however, a complex motion that requires the coordination of many degrees of freedom while it is also inherently unstable and sensitive to disturbances. The balance of a walking biped has to be constantly maintained. The most effective ways of controlling balance are well timed and placed recovery steps — capture steps — that absorb the expense momentum gained from a push or a stumble. We present a bipedal gait generation framework that utilizes step timing and foot placement techniques in order to recover the balance of a biped even after strong disturbances. Our framework modifies the next footstep location instantly when responding to a disturbance and generates controllable omnidirectional walking using only very little sensing and computational power. We exploit the open-loop stability of a central pattern generate...

2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2015

Bipedal walking is a complex and dynamic wholebody motion with balance constraints. Due to the in... more Bipedal walking is a complex and dynamic wholebody motion with balance constraints. Due to the inherently unstable inverted pendulum-like dynamics of walking, the design of robust walking controllers proved to be particularly challenging. While a controller could potentially be learned with a robot in the loop, the destructive nature of losing balance and the impracticality of a high number of repetitions render most existing learning methods unsuitable for an online learning setting with real hardware. We propose a model-driven learning method that enables a humanoid robot to quickly learn how to maintain its balance. We bootstrap the learning process with a central pattern generator for stepping motions that abstracts from the complexity of the walking motion and simplifies the problem setting to the learning of a small number of leg swing amplitude parameters. A simple physical model that represents the dominant dynamics of bipedal walking estimates an approximate gradient and suggests how to modify the swing amplitude to restore balance. In experiments with a real robot, we show that only a few failed steps are sufficient for our biped to learn strong push recovery skills in the sagittal direction.

KI - Künstliche Intelligenz, 2015

Bipedal walking is a complex whole-body motion with inherently unstable dynamics that makes the d... more Bipedal walking is a complex whole-body motion with inherently unstable dynamics that makes the design of a robust controller particularly challenging. While a walk controller could potentially be learned with the hardware in the loop, the destructive nature of exploratory motions and the impracticality of a high number of required repetitions render most of the existing machine learning methods unsuitable for an online learning setting with real hardware. In a project in the DFG Priority Programme Autonomous Learning, we are investigating ways of bootstrapping the learning process with basic walking skills and enabling a humanoid robot to autonomously learn how to control its balance during walking.

In recent years, humanoid soccer robots improved considerably. Elementary soccer skills, such as ... more In recent years, humanoid soccer robots improved considerably. Elementary soccer skills, such as bipedal walking, visual perception, and collision avoidance have matured enough to provide for dynamic and exciting soccer games. While the elementary skills still remain hot research topics, it is time to move forward and address higher level skills, such as motion planning and team play. In this work, we present a new method to generate ball approach trajectories by planning footstep sequences offline and training an online policy to meet the real time requirements of embedded systems with low computational power, as typically used for soccer robots. We compare the results with our current reactive behavior that was used in the last RoboCup competitions and show the improvements we achieved.

Bipedal walkers are difficult to control, inherently unstable systems. Besides the complexity of ... more Bipedal walkers are difficult to control, inherently unstable systems. Besides the complexity of the walking motion itself, the balance of the robot constantly has to be maintained with good foot placements and other disturbance-rejection strategies. In this work, we are presenting a new, closed-loop control approach that addresses both, the problem of complexity and the challenge of maintaining balance during walking. We decouple walking motion from balance and combine them in a hierarchical framework allowing a foot placement-based balance regulator to control the timing and footstep coordinates of central pattern-generated stepping motions. Furthermore, we decompose the balance controller into three simple, independent modules that compute suitable estimates of timing and sagittal and lateral coordinates for the next footstep to maintain a nominal center of mass trajectory. We implemented the timing and the lateral step size components using the equations of a parameterized versi...

Proceedings - IEEE International Conference on Robotics and Automation

Planning dynamic motion is computationally demanding and thus can hardly be done in real-time onb... more Planning dynamic motion is computationally demanding and thus can hardly be done in real-time onboard robots. In this paper, we present an analytic approximation to predict the dynamic state of wheeled robots with non-holonomic constraints, given a start state and a sequence of piecewise constant controls. Our approximations are accurate and fast to calculate. They can be used to replace numerical integrators in kinodynamic planning algorithms. The predictions are differentiable and allow us to utilize gradient descent methods to solve the inverse dynamics as well and generate trajectories connecting arbitrary points in state space.

Maintaining balance in the presence of disturbances is cru-cial for bipedal robots. In this paper... more Maintaining balance in the presence of disturbances is cru-cial for bipedal robots. In this paper, we focus on the lateral motion component. In order to attain disturbance rejection and to quickly re-cover balance, we combine three different control approaches. As a prin-cipal building block, we generate center of mass trajectories with a lin-ear model predictive controller that takes scheduled footsteps into ac-count. Strong disturbances generate unexpected angular momenta that can compromise stability. A second control layer extends the underlying preview controller with two recovery strategies that modify the planned CoM trajectories to dampen the rotational velocity of the robot and adapt the timing of the steps according to the expected orbital energy of CoM trajectories at support exchange. Experiments with a real Nao robot show that the system is able to recover from lateral disturbances as long as the robot does not tip over the current support leg.

Over the past few years, soccer-playing humanoid robots advanced significantly. Elementary skills... more Over the past few years, soccer-playing humanoid robots advanced significantly. Elementary skills, such as bipedal walking, visual perception, and collision avoidance have matured enough to allow for dynamic and exciting games. In this paper, team NimbRo TeenSize, the winner of the RoboCup 2012 Best Humanoid Award, presents its robotic platform and its approaches to perception and behavior control.

Bipedal walking is one of the most essential skills required to play soccer with humanoid robots.... more Bipedal walking is one of the most essential skills required to play soccer with humanoid robots. Superior walking speed and stability often gives teams the winning edge when their robots are the first at the ball, maintain ball control, and drive the ball towards the opponent goal with sure feet. In this contribution, we present an implementation of our Capture Step Framework on a real soccer robot, and show robust omnidirectional walking. The robot not only manages to locomote on an even surface, but can also cope with various disturbances, such as pushes, collisions, and stepping on the feet of an opponent. The actuation is compliant and the robot walks with stretched knees.

Most of the research related to the topic of falling strategies considers falling to be an unavoi... more Most of the research related to the topic of falling strategies considers falling to be an unavoidable part of bipedal walking and is focused on developing strategies to avoid falls and to minimize mechanical damage. We take an alternative point of view and regard falling as a means to an end. We present our falling strategy for the specific case of a robot soccer goalie that deliberately jumps in front of a moving ball to prevent it from rolling into the goal. The jump decision is based on observed ball position, speed and direction of movement. We show how we implement a targeted falling into the appropriate direction, minimize the time from the jump decision to ground impact, and what solutions we developed to prevent mechanical damage. The presented falling technique was used in RoboCup Humanoid KidSize and TeenSize competitions and proved to be essential for winning.

Due to the high complexity of the humanoid body, and its inherently unstable inverted pendulum-li... more Due to the high complexity of the humanoid body, and its inherently unstable inverted pendulum-like dynamics, the development of a robust and versatile walking controller proves to be a difficult task. Using machine learning algorithms with hardware in the loop is a promising way of achieving balanced and dynamic gaits. In this work, we propose an online learning technique that learns how to step onto a reference footstep location while maintaining the balance of a bipedal walker in the presence of disturbances. The ability to step with the help of a parametrized motion generator simplifies the learning problem to the low-dimensional space of footstep coordinates. To quickly adapt the produced step sizes from learned experience, we update an online-capable function ap-proximator with a pendulum-cart motivated gradient function that incorporates the trade-off between maintaining balance and stepping onto a desired location. While our method is able to robustly learn suitable footstep...

Most of the research related to the topic of falling strategies considers falling to be an unavoi... more Most of the research related to the topic of falling strategies considers falling to be an unavoidable part of bipedal walking and is focused on developing strategies to avoid falls and to minimize mechanical damage. We take an alternative point of view and regard falling as a means to an end. We present our falling strategy for the specific case of a robot soccer goalie that deliberately jumps in front of a moving ball to prevent it from rolling into the goal. The jump decision is based on observed ball position, speed and direction of movement. We show how we implement a targeted falling into the appropriate direction, minimize the time from the jump decision to ground impact, and what solutions we developed to prevent mechanical damage. The presented falling technique was used in RoboCup Humanoid KidSize and TeenSize competitions and proved to be essential for winning. I. INTRODUCTION Falling is an inevitable part of bipedal walking, especially in dynamic environments such as rob...

The detection of planar surfaces in a point cloud is a popular technique for the extraction of dr... more The detection of planar surfaces in a point cloud is a popular technique for the extraction of drivable or walkable surfaces and for tabletop segmentation. Unfortunately, RGBD sensors are quite noisy and provide incomplete data, which makes the extraction of surfaces more challenging. Also, it is desirable to process the point cloud data in real time, which at a rate of approximately 30 Hz, leaves only a small amount of computation time per frame. We have already developed a real time-capable plane segmentation method [1] that exploits the organized structure of RGB-D point clouds in order to implement a computationally efficient region growing algorithm. It uses the point-plane distance to assign points to their segments rather than inherently unreliable surface normals. Now we are presenting an improvement where we adapt thresholds and other parameters of our algorithm to the measured depth in order to account for an increasing scatter of the points at larger distances from the ca...

Foresighted navigation is an essential skill for robots to rise from rigid factory floor installa... more Foresighted navigation is an essential skill for robots to rise from rigid factory floor installations to much more versatile mobile robots that partake in our every day environment. The current state of the art that provides this mobility to some extent is the Dynamic Window Approach combined with a global start-to-target path planner. However, neither the Dynamic Window Approach nor the path planner are equipped to predict the motion of other objects in the environment. We propose a change in the Dynamic Window Approach—a dynamic collision model—that is capable of predicting future collisions with the environment, even when objects are in motion. We show in our experiments that our new way of computing the Dynamic Window Approach excels in navigational fitness in dynamic environments while being computationally efficient.

2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017

We present a simple and efficient mathematical framework suitable for generating motion in the co... more We present a simple and efficient mathematical framework suitable for generating motion in the context of a variety of robotic motion tasks ranging from low-level motor control up to high-level locomotion planning. Our concept is based on a one-dimensional second-order model that allows analytic computation of its inverse dynamics while respecting physical constraints. This makes it a particularly useful tool for tasks that are expressed only as a start and goal state, such as animation key frames or way points in path planning. By means of time synchronization, the model extends easily to an arbitrary number of dimensions in a way that the target is reached in all dimensions at the same time. The framework excels in terms of execution time, which lies in the microsecond range even for high-dimensional trajectory generation tasks. We demonstrate our method in two different settings — full-body trajectory generation and path planning — and show its benefits in comparison with current...

Bipedal walking is one of the most essential skills in humanoid robot soccer. A stable and fast g... more Bipedal walking is one of the most essential skills in humanoid robot soccer. A stable and fast gait gives teams a winning edge when their robots are the first at the ball, maintain ball control with sure feet, and drive the ball decisively towards the opponent goal. The most successful teams in the Humanoid League are typically characterized by reliably walking robots. In this contribution, we describe the omnidirectional gait of team NimbRo, one of the most successful robot soccer teams in the history of RoboCup. The walk algorithm is open-loop and model-free. It is based on highly configurable, centralpattern-generated rhythmic motion signals and combines well with a compliant servo setting to achieve a relatively high level of self-stability. We discuss the advantages of this approach in comparison with methods of other successful teams and support our argumentation with experimental results.

ArXiv, 2018

Over the past few years, a number of successful humanoid platforms have been developed, including... more Over the past few years, a number of successful humanoid platforms have been developed, including the Nao and the DARwIn-OP, both of which are used by many research groups for the investigation of bipedal walking, full-body motions, and human-robot interaction. The NimbRo-OP is an open humanoid platform under development by team NimbRo of the University of Bonn. Significantly larger than the two aforementioned humanoids, this platform has the potential to interact with a more human-scale environment. This paper describes a software framework for the NimbRo-OP that is based on the Robot Operating System (ROS) middleware. The software provides functionality for hardware abstraction, visual perception, and behavior generation, and has been used to implement basic soccer skills. These were demonstrated at RoboCup 2013, as part of the winning team of the Humanoid League competition.

2021 IEEE International Conference on Robotics and Automation (ICRA)

The segmentation of a point cloud into planar primitives is a popular approach to first-line scen... more The segmentation of a point cloud into planar primitives is a popular approach to first-line scene interpretation and is particularly useful in mobile robotics for the extraction of drivable or walkable surfaces and for tabletop segmentation for manipulation purposes. Unfortunately, the planar segmentation task becomes particularly challenging when the point clouds are obtained from an inherently noisy, robot-mounted sensor that is often in motion, therefor requiring real time processing capabilities. We present a real time-capable plane segmentation technique based on a region growing algorithm that exploits the organized structure of point clouds obtained from RGB-D sensors. In order to counteract the sensor noise, we invest into careful selection of seeds that start the region growing and avoid the computation of surface normals whenever possible. We implemented our algorithm in C++ and thoroughly tested it in both simulated and real-world environments where we are able to compare our approach against existing state-of-the-art methods implemented in the Point Cloud Library. The experiments presented here suggest that our approach is accurate and fast, even in the presence of considerable sensor noise.

2021 IEEE International Conference on Robotics and Automation (ICRA)

Footstep planning is the dominating approach when it comes to controlling the walk of a humanoid ... more Footstep planning is the dominating approach when it comes to controlling the walk of a humanoid robot, even though a footstep plan is expensive to compute. The most prominent proposals typically spend up to a few seconds of computation time and output a sequence of up to 30 steps all the way to the goal. This way, footstep planning is applicable only in static environments where nothing changes after a plan has been computed. Since uncontrolled environments present challenges such as unforeseen motion of other objects and unexpected disturbances to balance, fast replanning of a footstep plan while the robot is in motion is highly desirable. We present a new way of fast footstep planning-Aborting A*-which is able to guarantee a replanning rate of 50 Hz by aborting an A* search before completion. We make aborting possible by using a novel, obstacle-aware heuristic function that lays out rotate-translaterotate motions along the shortest path to the goal, enabling us to stop the planning progress prematurely with a target-oriented solution at any time during the search, even after only a few nodes have been expanded. We show in our experiments that despite the bounded computation time, our planner computes good results and does not get stuck in local minima.

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

With the advent of polygonal maps finding their way into the navigational software of mobile robo... more With the advent of polygonal maps finding their way into the navigational software of mobile robots, the Visibility Graph can be used to search for the shortest collisionfree path. The nature of the Visibility Graph-based shortest path algorithms is such that first the entire graph is computed in a relatively time-consuming manner. Then, the graph can be searched efficiently any number of times for varying start and target state combinations with the A* or the Dijkstra algorithm. However, real-world environments are typically too dynamic for a map to remain valid for a long time. With the goal of obtaining the shortest path quickly in an ever changing environment, we introduce a rapid path finding algorithm-Minimal Construct-that discovers only a necessary portion of the Visibility Graph around the obstacles that actually get in the way. Collision tests are computed during an A* search only for lines that seem heuristically promising. This way, shortest paths can be found much faster than with a state-of-the-art Visibility Graph algorithm and as our experiments show, even grid-based A* searches are outperformed in most cases with the added benefit of smoother and shorter paths.

International Journal of Humanoid Robotics

Stable bipedal walking is a key prerequisite for humanoid robots to reach their potential of bein... more Stable bipedal walking is a key prerequisite for humanoid robots to reach their potential of being versatile helpers in our everyday environments. Bipedal walking is, however, a complex motion that requires the coordination of many degrees of freedom while it is also inherently unstable and sensitive to disturbances. The balance of a walking biped has to be constantly maintained. The most effective ways of controlling balance are well timed and placed recovery steps — capture steps — that absorb the expense momentum gained from a push or a stumble. We present a bipedal gait generation framework that utilizes step timing and foot placement techniques in order to recover the balance of a biped even after strong disturbances. Our framework modifies the next footstep location instantly when responding to a disturbance and generates controllable omnidirectional walking using only very little sensing and computational power. We exploit the open-loop stability of a central pattern generate...

2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2015

Bipedal walking is a complex and dynamic wholebody motion with balance constraints. Due to the in... more Bipedal walking is a complex and dynamic wholebody motion with balance constraints. Due to the inherently unstable inverted pendulum-like dynamics of walking, the design of robust walking controllers proved to be particularly challenging. While a controller could potentially be learned with a robot in the loop, the destructive nature of losing balance and the impracticality of a high number of repetitions render most existing learning methods unsuitable for an online learning setting with real hardware. We propose a model-driven learning method that enables a humanoid robot to quickly learn how to maintain its balance. We bootstrap the learning process with a central pattern generator for stepping motions that abstracts from the complexity of the walking motion and simplifies the problem setting to the learning of a small number of leg swing amplitude parameters. A simple physical model that represents the dominant dynamics of bipedal walking estimates an approximate gradient and suggests how to modify the swing amplitude to restore balance. In experiments with a real robot, we show that only a few failed steps are sufficient for our biped to learn strong push recovery skills in the sagittal direction.

KI - Künstliche Intelligenz, 2015

Bipedal walking is a complex whole-body motion with inherently unstable dynamics that makes the d... more Bipedal walking is a complex whole-body motion with inherently unstable dynamics that makes the design of a robust controller particularly challenging. While a walk controller could potentially be learned with the hardware in the loop, the destructive nature of exploratory motions and the impracticality of a high number of required repetitions render most of the existing machine learning methods unsuitable for an online learning setting with real hardware. In a project in the DFG Priority Programme Autonomous Learning, we are investigating ways of bootstrapping the learning process with basic walking skills and enabling a humanoid robot to autonomously learn how to control its balance during walking.

In recent years, humanoid soccer robots improved considerably. Elementary soccer skills, such as ... more In recent years, humanoid soccer robots improved considerably. Elementary soccer skills, such as bipedal walking, visual perception, and collision avoidance have matured enough to provide for dynamic and exciting soccer games. While the elementary skills still remain hot research topics, it is time to move forward and address higher level skills, such as motion planning and team play. In this work, we present a new method to generate ball approach trajectories by planning footstep sequences offline and training an online policy to meet the real time requirements of embedded systems with low computational power, as typically used for soccer robots. We compare the results with our current reactive behavior that was used in the last RoboCup competitions and show the improvements we achieved.

Bipedal walkers are difficult to control, inherently unstable systems. Besides the complexity of ... more Bipedal walkers are difficult to control, inherently unstable systems. Besides the complexity of the walking motion itself, the balance of the robot constantly has to be maintained with good foot placements and other disturbance-rejection strategies. In this work, we are presenting a new, closed-loop control approach that addresses both, the problem of complexity and the challenge of maintaining balance during walking. We decouple walking motion from balance and combine them in a hierarchical framework allowing a foot placement-based balance regulator to control the timing and footstep coordinates of central pattern-generated stepping motions. Furthermore, we decompose the balance controller into three simple, independent modules that compute suitable estimates of timing and sagittal and lateral coordinates for the next footstep to maintain a nominal center of mass trajectory. We implemented the timing and the lateral step size components using the equations of a parameterized versi...

Proceedings - IEEE International Conference on Robotics and Automation

Planning dynamic motion is computationally demanding and thus can hardly be done in real-time onb... more Planning dynamic motion is computationally demanding and thus can hardly be done in real-time onboard robots. In this paper, we present an analytic approximation to predict the dynamic state of wheeled robots with non-holonomic constraints, given a start state and a sequence of piecewise constant controls. Our approximations are accurate and fast to calculate. They can be used to replace numerical integrators in kinodynamic planning algorithms. The predictions are differentiable and allow us to utilize gradient descent methods to solve the inverse dynamics as well and generate trajectories connecting arbitrary points in state space.

Maintaining balance in the presence of disturbances is cru-cial for bipedal robots. In this paper... more Maintaining balance in the presence of disturbances is cru-cial for bipedal robots. In this paper, we focus on the lateral motion component. In order to attain disturbance rejection and to quickly re-cover balance, we combine three different control approaches. As a prin-cipal building block, we generate center of mass trajectories with a lin-ear model predictive controller that takes scheduled footsteps into ac-count. Strong disturbances generate unexpected angular momenta that can compromise stability. A second control layer extends the underlying preview controller with two recovery strategies that modify the planned CoM trajectories to dampen the rotational velocity of the robot and adapt the timing of the steps according to the expected orbital energy of CoM trajectories at support exchange. Experiments with a real Nao robot show that the system is able to recover from lateral disturbances as long as the robot does not tip over the current support leg.

Over the past few years, soccer-playing humanoid robots advanced significantly. Elementary skills... more Over the past few years, soccer-playing humanoid robots advanced significantly. Elementary skills, such as bipedal walking, visual perception, and collision avoidance have matured enough to allow for dynamic and exciting games. In this paper, team NimbRo TeenSize, the winner of the RoboCup 2012 Best Humanoid Award, presents its robotic platform and its approaches to perception and behavior control.

Bipedal walking is one of the most essential skills required to play soccer with humanoid robots.... more Bipedal walking is one of the most essential skills required to play soccer with humanoid robots. Superior walking speed and stability often gives teams the winning edge when their robots are the first at the ball, maintain ball control, and drive the ball towards the opponent goal with sure feet. In this contribution, we present an implementation of our Capture Step Framework on a real soccer robot, and show robust omnidirectional walking. The robot not only manages to locomote on an even surface, but can also cope with various disturbances, such as pushes, collisions, and stepping on the feet of an opponent. The actuation is compliant and the robot walks with stretched knees.

Most of the research related to the topic of falling strategies considers falling to be an unavoi... more Most of the research related to the topic of falling strategies considers falling to be an unavoidable part of bipedal walking and is focused on developing strategies to avoid falls and to minimize mechanical damage. We take an alternative point of view and regard falling as a means to an end. We present our falling strategy for the specific case of a robot soccer goalie that deliberately jumps in front of a moving ball to prevent it from rolling into the goal. The jump decision is based on observed ball position, speed and direction of movement. We show how we implement a targeted falling into the appropriate direction, minimize the time from the jump decision to ground impact, and what solutions we developed to prevent mechanical damage. The presented falling technique was used in RoboCup Humanoid KidSize and TeenSize competitions and proved to be essential for winning.

Due to the high complexity of the humanoid body, and its inherently unstable inverted pendulum-li... more Due to the high complexity of the humanoid body, and its inherently unstable inverted pendulum-like dynamics, the development of a robust and versatile walking controller proves to be a difficult task. Using machine learning algorithms with hardware in the loop is a promising way of achieving balanced and dynamic gaits. In this work, we propose an online learning technique that learns how to step onto a reference footstep location while maintaining the balance of a bipedal walker in the presence of disturbances. The ability to step with the help of a parametrized motion generator simplifies the learning problem to the low-dimensional space of footstep coordinates. To quickly adapt the produced step sizes from learned experience, we update an online-capable function ap-proximator with a pendulum-cart motivated gradient function that incorporates the trade-off between maintaining balance and stepping onto a desired location. While our method is able to robustly learn suitable footstep...