Petar Kormushev

• PhD in Computational Intelligence
• Lecturer at Imperial College London

Knee-less bipedal robots like SLIDER have the advantage of ultra-lightweight legs and improved walking energy efficiency compared to traditional humanoid robots. In this paper, we firstly introduce an improved hardware design of the bipedal robot SLIDER with new line-feet and more optimized mass distribution which enables higher locomotion speeds....

Physical manipulation of garments is often crucial when performing fabric-related tasks, such as hanging garments. However, due to the deformable nature of fabrics, these operations remain a significant challenge for robots in household, healthcare, and industrial environments. In this paper, we propose GraphGarment, a novel approach that models ga...

End-to-end robot learning, particularly for long-horizon tasks, often results in unpredictable outcomes and poor generalization. To address these challenges, we propose a novel Therblig-Based Backbone Framework (TBBF) as a fundamental structure to enhance interpretability, data efficiency, and generalization in robotic systems. TBBF utilizes expe...

Myoelectric prosthetic hands are typically controlled to move between discrete positions and do not provide sensory feedback to the user. In this work, we present and evaluate a closed-loop, continuous myoelectric prosthetic hand controller, that can continuously control the position of multiple degrees of freedom of a prosthesis while rendering pr...

Despite decades of research and development, myoelectric prosthetic hands lack functionality and are often rejected by users. This lack in functionality can be partially attributed to the widely accepted anthropomorphic design ideology in the field; attempting to replicate human hand form and function despite severe limitations in control and sensi...

Myoelectric prosthetic hands are typically controlled to move between discrete positions and do not provide sensory feedback to the user. In this work, we present and evaluate a closed-loop, continuous myoelectric prosthetic hand controller, that can continuously control the position of multiple degrees of freedom of a prosthesis while rendering pr...

Despite decades of research and development, myoelectric prosthetic hands lack functionality and are often rejected by users. This lack in functionality can be attributed to the widely accepted anthropomorphic design ideology in the field; attempting to replicate human hand form and function despite severe limitations in control and sensing technol...

Robotic manipulation is essential for the widespread adoption of robots in industrial and home settings and has long been a focus within the robotics community. Advances in artificial intelligence have introduced promising learning-based methods to address this challenge, with imitation learning emerging as particularly effective. However, efficien...

End-to-end robot learning, particularly for long-horizon tasks, often results in unpredictable outcomes and poor generalization. To address these challenges, we propose a novel Therblig-based Backbone Framework (TBBF) to enhance robot task understanding and transferability. This framework uses therbligs (basic action elements) as the backbone to de...

Human hands are able to grasp a wide range of object sizes, shapes, and weights, achieved via reshaping and altering their apparent grasping stiffness between compliant power and rigid precision. Achieving similar versatility in robotic hands remains a challenge, which has often been addressed by adding extra controllable degrees of freedom, tactil...

To achieve highly dynamic jumps with legged robots, it is essential to control the rotational dynamics of the robot. In this paper, we aim to improve the jumping performance by proposing a unified model for planning highly dynamic jumps that can approximately model the centroidal inertia. This model abstracts the robot as a single rigid body for th...

In minimally invasive robotic surgery, the surgical instrument is usually inserted inside the patient’s body through a small incision, which acts as a remote center of motion (RCM). Serial-link manipulators can be used as macro robots on which microsurgical robotic instruments are mounted to increase the number of degrees of freedom of the system a...

Prosthetic hand control research typically focuses on developing increasingly complex controllers to achieve diminishing returns in pattern recognition of muscle activity signals, making models less suitable for user calibration. Some works have investigated transfer learning to alleviate this, but such approaches increase model size dramatically-t...

Robots for minimally invasive surgery are becoming more and more complex, due to miniaturization and flexibility requirements. The vast majority of surgical robots are tendon-driven and this, along with the complex design, causes high nonlinearities in the system which are difficult to model analytically. In this work we analyse how incorporating a...

Due to the increasing complexity of robotic structures, modelling robots is becoming more and more challenging, and analytical models are very difficult to build. Machine learning approaches have shown great capabilities in learning complex mapping and have widely been used in robot model learning and control. Generally, the inverse kinematics is d...

This paper presents the implementation and evaluation of three specific, yet complementary, mechanisms of haptic feedback—namely, normal displacement, tangential position, and vibration—to render, at a finger-level, aspects of touch and proprioception from a prosthetic hand without specialised sensors. This feedback is executed by an armband worn a...

We present an online optimization algorithm which enables bipedal robots to blindly walk over various kinds of uneven terrains while resisting pushes. The proposed optimization algorithm performs high-level motion planning of footstep locations and center-of-mass height variations using the decoupled actuated spring-loaded inverted pendulum (aSLIP)...

Model predictive control is a widely used optimal control method for robot path planning and obstacle avoidance. This control method, however, requires a system model to optimize control over a finite time horizon and possible trajectories. Certain types of robots, such as soft robots, continuum robots, and transforming robots, can be challenging t...

Robot design is a major component in robotics, as it allows building robots capable of performing properly in given tasks. However, designing a robot with multiple types of parameters and constraints and defining an optimization function analytically for the robot design problem may be intractable or even impossible. Therefore black-box optimizatio...

Most state-of-the-art bipedal robots are designed to be anthropomorphic, and therefore possess articulated legs with knees. Whilst this facilitates smoother, human-like locomotion, there are implementation issues that make walking with straight legs difficult. Many robots have to move with a constant bend in the legs to avoid a singularity occurrin...

Accurate robot kinematic modelling is a major component for autonomous robot control to guarantee safety and precision during task execution. In surgical robotics complex robotic structures and actuation mechanisms are generally employed, therefore machine learning techniques can be adopted to build the model of the robot. Probabilistic neural netw...

Muscle-actuated control is a research topic of interest spanning different fields, in particular biomechanics, robotics and graphics. This type of control is particularly challenging because models are often overactuated, and dynamics are delayed and non-linear. It is however a very well tested and tuned actuation model that has undergone millions...

Model learning is a crucial aspect of robotics as it enables the use of traditional and consolidated model-based controllers to perform desired motion tasks. However, due to the increasing complexity of robotic structures, modelling robots is becoming more and more challenging, and analytical models are very difficult to build, particularly for red...

We present a highly reactive controller which enables bipedal robots to blindly walk over various kinds of uneven terrains while resisting pushes. The high level motion planner does fast online optimization for footstep locations and Center of Mass (CoM) height using the decoupled actuated Spring Loaded Inverted Pendulum (aSLIP) model. The decouple...

To achieve highly dynamic jumps of legged robots, it is essential to control the rotational dynamics of the robot. In this paper, we aim to improve the jumping performance by proposing a unified model for planning highly dynamic jumps that can approximately model the centroidal inertia. This model abstracts the robot as a single rigid body for the...

Despite the fact that a large number of research studies have been conducted in the field of search and rescue robotics, significantly little attention has been given to the development of rescue robots capable of performing physical rescue interventions, including loading and transporting victims to a safe zone—i.e., casualty extraction tasks. The...

Robots have been predominantly controlled using conventional control methods that require prior knowledge of the robots’ kinematic and dynamic models. These controllers can be challenging to tune and cannot directly adapt to changes in kinematic structure or dynamic properties. On the other hand, model-learning controllers can overcome such challen...

Neuroevolution is an alternative to gradient-based optimisation that has the potential to avoid local minima and allows parallelisation. The main limiting factor is that usually it does not scale well with parameter space dimensionality. Inspired by recent work examining neural network intrinsic dimension and loss landscapes, we hypothesise that th...

Stiffness modulation in walking is critical to maintain static/dynamic stability as well as to minimize energy consumption and impact damage. However, optimal, or even functional, stiffness parameterization remains unresolved in legged robotics. We introduce an architecture for stiffness control utilising a bioinspired robotic limb consisting of a...

In recent years, several robots have been developed and deployed to perform casualty extraction tasks. However, the majority of these robots are overly complex, and require teleoperation via either a skilled operator or a specialised device, and often the operator must be present at the scene to navigate safely around the casualty. Instead, improvi...

In order to guarantee precision and safety in robotic surgery, accurate models of the robot and proper control strategies are needed. Bayesian Neural Networks (BNN) are capable of learning complex models and provide information about the uncertainties of the learned system. Model Predictive Control (MPC) is a reliable control strategy to ensure opt...

Diversity-based approaches have recently gained popularity as an alternative paradigm to performance-based policy search. A popular approach from this family, Quality-Diversity (QD), maintains a collection of high-performing policies separated in the diversity-metric space, defined based on policies' rollout behaviours. When policies are parameteri...

Action-value estimation is a critical component of many reinforcement learning (RL) methods whereby sample complexity relies heavily on how fast a good estimator for action value can be learned. By viewing this problem through the lens of representation learning, good representations of both state and action can facilitate action-value estimation....

Conventional control of robotic manipulators requires prior knowledge of their kinematic structure. Model-learning controllers have the advantage of being able to control robots without requiring a complete kinematic model and work well in less structured environments. Our recently proposed Encoderless controller has shown promising ability to cont...

Accurate kinematic models are essential for effective control of surgical robots. For tendon driven robots, which are common for minimally invasive surgery, the high nonlinearities in the transmission make modelling complex. Machine learning techniques are a preferred approach to tackle this problem. However, surgical environments are rarely struct...

Online footstep planning is essential for bipedal walking robots to be able to walk in the presence of disturbances. Until recently this has been achieved by only optimizing the placement of the footstep, keeping the duration of the step constant. In this paper we introduce a footstep planner capable of optimizing footstep placement and timing in r...

We introduce Robot DE NIRO, an autonomous, collaborative, humanoid robot for mobile manipulation. We built DE NIRO to perform a wide variety of manipulation behaviors, with a focus on pick-and-place tasks. DE NIRO is designed to be used in a domestic environment, especially in support of caregivers working with the elderly. Given this design focus,...

Being able to reach any desired location in the environment can be a valuable asset for an agent. Learning a policy to navigate between all pairs of states individually is often not feasible. An all-goals updating algorithm uses each transition to learn Q-values towards all goals simultaneously and off-policy. However the expensive numerous updates...

Objective: Elucidating the role of structural mechanisms in the knee can improve joint surgeries, rehabilitation, and understanding of biped locomotion. Identification of key features, however, is challenging due to limitations in simulation and in-vivo studies. In particular the coupling of the patello-femoral and tibio-femoral joints with ligame...

Learning complex physical tasks via trial-and-error is still challenging for high-degree-of-freedom robots. Greatest challenges are devising a suitable objective function that defines the task, and the high sample complexity of learning the task. We propose a novel active learning framework, consisting of decoupled task model and exploration compon...

ResQbot 2.0: A Mobile Stretcher Robot with Neck Securing Device for Safe Casualty Extraction

This extended abstract has been presented in the Late Breaking Results of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

The aim of our study is to develop a mobile rescue robot that could assist first responders when saving casualties from a danger area by performing a casualty extraction procedure, whilst ensuring that no additional injury is caused by the operation and no additional lives are put at risk. Specifically, in this project, we significantly improved th...

This paper addresses the problem of human body detection---particularly a human body lying on the ground (a.k.a. casualty)---using point cloud data. This ability to detect a casualty is one of the most important features of mobile rescue robots, in order for them to be able to operate autonomously. We propose a deep-learning-based casualty detectio...

This paper presents a novel autonomous air-hockey playing collaborative robot (cobot) that provides human-like gameplay against human opponents. Vision-based Bayesian tracking of the puck and striker are used in an Analytic Hierarchy Process (AHP)-based probabilistic tactical layer for high-speed perception. The tactical layer provides commands for...

We consider the generic approach of using an experience memory to help exploration by adapting a restart distribution. That is, given the capacity to reset the state with those corresponding to the agent's past observations, we help exploration by promoting faster state-space coverage via restarting the agent from a more diverse set of initial stat...

While robotics has made significant advances in perception, planning and control in recent decades, the vast majority of tasks easily completed by a human, especially acting in dynamic, unstructured environments, are far from being autonomously performed by a robot. Teleoperation, remotely controlling a slave robot by a human operator, can be a rea...

Modern humanoid robots include not only active compliance but also passive compliance. Apart from improved safety and dependability, availability of passive elements, such as springs, opens up new possibilities for improving the energy efficiency. With this in mind, this paper addresses the challenging open problem of exploiting the passive complia...

Social assistance robots in health and elderly care have the potential to support and ease human lives. Given the macrosocial trends of aging and long-lived populations, robotics-based care research mainly focused on helping the elderly live independently. In this paper, we introduce Robot DE NIRO, a research platform that aims to support the suppo...

Goal-oriented learning has become a core concept in reinforcement learning (RL), extending the reward signal as a sole way to define tasks. However, as parameterizing value functions with goals increases the learning complexity, efficiently reusing past experience to update estimates towards several goals at once becomes desirable but usually requi...

One of the most important features of mobile rescue robots is the ability to autonomously detect casualties, i.e. human bodies, which are usually lying on the ground. This paper proposes a novel method for autonomously detecting casualties lying on the ground using obtained 3D point-cloud data from an on-board sensor, such as an RGB-D camera or a 3...

One of the most important features of mobile rescue robots is the ability to autonomously detect casualties, i.e. human bodies, which are usually lying on the ground. This paper proposes a novel method for autonomously detecting casualties lying on the ground using obtained 3D point-cloud data from an on-board sensor, such as an RGB-D camera or a 3...

Exploration is a difficult challenge in reinforcement learning and even recent state-of-the art curiosity-based methods rely on the simple epsilon-greedy strategy to generate novelty. We argue that pure random walks do not succeed to properly expand the exploration area in most environments and propose to replace single random action choices by ran...

In this work, we propose a novel mobile rescue robot equipped with an immersive stereoscopic teleperception and a teleoperation control. This robot is designed with the capability to perform safely a casualty-extraction procedure. We have built a proof-of-concept mobile rescue robot called ResQbot for the experimental platform. An approach called “...

In order to operate autonomously, mobile rescue robots need to be able to detect human casualties in disaster situations. In this paper, we propose a novel method for autonomous detection of casualties lying down on the ground based on point-cloud data. This data can be obtained from different sensors, such as an RGB-D camera or a 3D LIDAR sensor....

Performing search and rescue missions in disaster-struck environments is challenging. Despite the advances in the robotic search phase of the rescue missions, few works have been focused on the physical casualty extraction phase. In this work, we propose a mobile rescue robot that is capable of performing a safe casualty extraction routine. To perf...

Performing search and rescue missions in disaster-struck environments is challenging. Despite the advances in the robotic search phase of the rescue missions, few works have been focused on the physical casualty extraction phase. In this work, we propose a mobile rescue robot that is capable of performing a safe casualty extraction routine. To perf...

In reinforcement learning, it is common to let an agent interact with its environment for a fixed amount of time before resetting the environment and repeating the process in a series of episodes. The task that the agent has to learn can either be to maximize its performance over (i) that fixed period, or (ii) an indefinite period where time limits...

Discrete-action algorithms have been central to numerous recent successes of deep reinforcement learning. However, applying these algorithms to high-dimensional action tasks requires tackling the combinatorial increase of the number of possible actions with the number of action dimensions. This problem is further exacerbated for continuous-action t...

Discrete-action algorithms have been central to numerous recent successes of deep reinforcement learning. However, applying these algorithms to high-dimensional action tasks requires tackling the combinatorial increase of the number of possible actions with the number of action dimensions. This problem is further exacerbated for continuous-action t...

Incremental progress in humanoid robot locomotion over the years has achieved important capabilities such as navigation over flat or uneven terrain, stepping over small obstacles and climbing stairs. However, the locomotion research has mostly been limited to using only bipedal gait and only foot contacts with the environment, using the upper body...

A novel approach for combined locomotion and manipulation (loco-manipulation) for an autonomous mobile rescue robot is proposed. Rescuing human victim due to natural or man-made disasters in potentially dangerous places poses huge challenges for both locomotion and manipulation and requires sophisticated rescue robots. Our research goal is to devel...

A novel skill learning approach is proposed that allows a robot to acquire human-like visuospatial skills for object manipulation tasks. Visuospatial skills are attained by observing spatial relationships among objects through demonstrations. The proposed Visuospatial Skill Learning (VSL) is a goal-based approach that focuses on achieving a desired...

Intervention autonomous underwater vehicles (I-AUVs) have the potential to open new avenues for the maintenance and monitoring of offshore subsea facilities in a cost-effective way. However, this requires challenging intervention operations to be carried out persistently, thus minimizing human supervision and ensuring a reliable vehicle behaviour u...

This paper presents modular dynamics for dual-arms, expressed in terms of the kinematics and dynamics of each of the stand-alone manipulators. The two arms are controlled as a single manipulator in the task space that is relative to the two end-effectors of the dual-arm robot. A modular relative Jacobian, derived from a previous work, is used which...

This paper presents some of the results of the EU-funded project PANDORA — Persistent Autonomy Through Learning Adaptation Observation and Re-planning. The project was three and a half years long and involved several organisations across Europe. The application domain is underwater inspection and intervention, a topic particularly interesting for t...

This chapter introduces Visuospatial Skill Learning (VSL), which is a novel interactive robot learning approach. VSL is based on visual perception that allows a robot to acquire new skills by observing a single demonstration while interacting with a tutor. The focus of VSL is placed on achieving a desired goal configuration of objects relative to a...

Autonomous robots are not very good at being autonomous. They work well in structured environments, but fail quickly in the real world facing uncertainty and dynamically changing conditions. In this chapter, we describe robot learning approaches that help to elevate robot autonomy to the next level, the so-called ‘persistent autonomy’. For a robot...

PANDORA is a EU FP7 project that is developing new computational methods to make underwater robots Persistently Autonomous, significantly reducing the frequency of assistance requests. The aim of the project is to extend the range of tasks that can be carried on autonomously and increase their complexity while reducing the need for operator assista...

This paper challenges the well-established assumption in robotics that in order to control a robot it is necessary to know its kinematic information, that is, the arrangement of links and joints, the link dimensions and the joint positions. We propose a kinematic-free robot control concept that does not require any prior kinematic knowledge. The co...

We propose a new algorithm capable of online regeneration of walking gait patterns. The algorithm uses a nonlinear optimization technique to find step parameters that will bring the robot from the present state to a desired state. It modifies online not only the footstep positions, but also the step timing in order to maintain dynamic stability dur...

A modular relative Jacobian is recently derived and is expressed in terms of the individual Jacobians of stand-alone manipulators. It includes a wrench transformation matrix, which was not shown in earlier expressions. This paper is an experimental extension of that recent work, which showed that at higher angular end-effector velocities the contri...

The implementation of autonomous intervention tasks with underwater vehicles is a non-trivial issue due to the challenging and dynamic conditions of the underwater medium (e.g., water current perturbations, water visibility). Likewise, it requires a significant programming effort each time that the vehicle must perform a different manipulation oper...

In this paper, a robot learning approach is proposed which integrates Visuospatial Skill Learning, Imitation Learning, and conventional planning methods. In our approach, the sensorimotor skills (i.e., actions) are learned through a learning from demonstration strategy. The sequence of performed actions is learned through demonstrations using Visu-...

Encoders have been an inseparable part of robots since the very beginning of modern robotics in the 1950s. As a result, the foundations of robot control are built on the concepts of kinematics and dynamics of articulated rigid bodies, which rely on explicitly measuring the robot configuration in terms of joint angles – done by encoders. In this pap...

Autonomous manipulation of objects requires reliable information on robot-object contact state. Underwater environments can adversely affect sensing modalities such as vision, making them unreliable. In this paper we investigate underwater robot-object contact perception between an autonomous underwater vehicle and a T-bar valve using a force/torqu...

Recent efforts in the field of intervention-autonomous underwater vehicles (I-AUVs) have started to show promising results in simple manipulation tasks. However, there is still a long way to go to reach the complexity of the tasks carried out by ROV pilots. This paper proposes an intervention framework based on parametric Learning by Demonstration...

It is essential for a successful completion of a robot object grasping and manipulation task to accurately sense the manipulated object’s pose. Typically, computer vision is used to obtain this information, but it may not be available or be reliable in certain situations. This paper presents a global optimisation method where tactile and force sens...