Data collection in robotics is spread across diverse hardware, and this variation will increase as new hardware is developed. Effective use of this growing body of data requires methods capable of learning from diverse robot embodiments. We consider the setting of training a policy using expert trajectories from a single robot arm (the source), and evaluating on a different robot arm for which no data was collected (the target). We present a data editing scheme termed Shadow, in which the robot during training and evaluation is replaced with a composite segmentation mask of the source and target robots. In this way, the input data distribution at train and test time match closely, enabling robust policy transfer to the new unseen robot while being far more data efficient than approaches that require co-training on large amounts of data from diverse embodiments. We demonstrate that an approach as simple as Shadow is effective both in simulation on varying tasks and robots, and on real robot hardware, where Shadow demonstrates an average of over 2x improvement in success rate compared to the strongest baseline.

Abstract-- In this paper, we consider an arbitrary number of joints and their arbitrary joint locations along the center line of a displacement-actuated continuum robot. To achieve this, we revisit the derivation of the Clarke transform leading to a formulation capable of considering arbitrary joint locations. The proposed modified Clarke transform opens new opportunities in mechanical design and algorithmic approaches beyond the current limiting dependency on symmetric arranged joint locations. By presenting an encoder-decoder architecture based on the Clarke transform, joint values between different robot designs can be transformed enabling the use of an analogous robot design and direct knowledge transfer. To demonstrate its versatility, applications of control and trajectory generation in simulation are presented, which can be easily integrated into an existing framework designed, for instance, for three symmetric arranged joints. Joint location and improved joint representation.

In indoor environments, multi-robot visual (RGB-D) mapping and exploration hold immense potential for application in domains such as domestic service and logistics, where deploying multiple robots in the same environment can significantly enhance efficiency. However, there are two primary challenges: (1) the "ghosting trail" effect, which occurs due to overlapping views of robots impacting the accuracy and quality of point cloud reconstruction, and (2) the oversight of visual reconstructions in selecting the most effective frontiers for exploration. Given these challenges are interrelated, we address them together by proposing a new semi-distributed framework (SPACE) for spatial cooperation in indoor environments that enables enhanced coverage and 3D mapping. SPACE leverages geometric techniques, including "mutual awareness" and a "dynamic robot filter," to overcome spatial mapping constraints. Additionally, we introduce a novel spatial frontier detection system and map merger, integrated with an adaptive frontier assigner for optimal coverage balancing the exploration and reconstruction objectives. In extensive ROS-Gazebo simulations, SPACE demonstrated superior performance over state-of-the-art approaches in both exploration and mapping metrics.

For tasks conducted in unknown environments with efficiency requirements, real-time navigation of multi-robot systems remains challenging due to unfamiliarity with surroundings.In this paper, we propose a novel multi-robot collaborative planning method that leverages the perception of different robots to intelligently select search directions and improve planning efficiency. Specifically, a foundational planner is employed to ensure reliable exploration towards targets in unknown environments and we introduce Graph Attention Architecture with Information Gain Weight(GIWT) to synthesizes the information from the target robot and its teammates to facilitate effective navigation around obstacles.In GIWT, after regionally encoding the relative positions of the robots along with their perceptual features, we compute the shared attention scores and incorporate the information gain obtained from neighboring robots as a supplementary weight. We design a corresponding expert data generation scheme to simulate real-world decision-making conditions for network training. Simulation experiments and real robot tests demonstrates that the proposed method significantly improves efficiency and enables collaborative planning for multiple robots. Our method achieves approximately 82% accuracy on the expert dataset and reduces the average path length by about 8% and 6% across two types of tasks compared to the fundamental planner in ROS tests, and a path length reduction of over 6% in real-world experiments.

Scaling up robot learning requires large and diverse datasets, and how to efficiently reuse collected data and transfer policies to new embodiments remains an open question. Emerging research such as the Open-X Embodiment (OXE) project has shown promise in leveraging skills by combining datasets including different robots. However, imbalances in the distribution of robot types and camera angles in many datasets make policies prone to overfit. To mitigate this issue, we propose RoVi-Aug, which leverages state-of-the-art image-to-image generative models to augment robot data by synthesizing demonstrations with different robots and camera views. Through extensive physical experiments, we show that, by training on robot- and viewpoint-augmented data, RoVi-Aug can zero-shot deploy on an unseen robot with significantly different camera angles. Compared to test-time adaptation algorithms such as Mirage, RoVi-Aug requires no extra processing at test time, does not assume known camera angles, and allows policy fine-tuning. Moreover, by co-training on both the original and augmented robot datasets, RoVi-Aug can learn multi-robot and multi-task policies, enabling more efficient transfer between robots and skills and improving success rates by up to 30%. Project website: https://rovi-aug.github.io.

The ability to reuse collected data and transfer trained policies between robots could alleviate the burden of additional data collection and training. While existing approaches such as pretraining plus finetuning and co-training show promise, they do not generalize to robots unseen in training. Focusing on common robot arms with similar workspaces and 2-jaw grippers, we investigate the feasibility of zero-shot transfer. Through simulation studies on 8 manipulation tasks, we find that state-based Cartesian control policies can successfully zero-shot transfer to a target robot after accounting for forward dynamics. To address robot visual disparities for vision-based policies, we introduce Mirage, which uses "cross-painting"--masking out the unseen target robot and inpainting the seen source robot--during execution in real time so that it appears to the policy as if the trained source robot were performing the task. Mirage applies to both first-person and third-person camera views and policies that take in both states and images as inputs or only images as inputs. Despite its simplicity, our extensive simulation and physical experiments provide strong evidence that Mirage can successfully zero-shot transfer between different robot arms and grippers with only minimal performance degradation on a variety of manipulation tasks such as picking, stacking, and assembly, significantly outperforming a generalist policy. Project website: https://robot-mirage.github.io/

For reliable autonomous robot navigation in urban settings, the robot must have the ability to identify semantically traversable terrains in the image based on the semantic understanding of the scene. This reasoning ability is based on semantic traversability, which is frequently achieved using semantic segmentation models fine-tuned on the testing domain. This fine-tuning process often involves manual data collection with the target robot and annotation by human labelers which is prohibitively expensive and unscalable. In this work, we present an effective methodology for training a semantic traversability estimator using egocentric videos and an automated annotation process. Egocentric videos are collected from a camera mounted on a pedestrian's chest. The dataset for training the semantic traversability estimator is then automatically generated by extracting semantically traversable regions in each video frame using a recent foundation model in image segmentation and its prompting technique. Extensive experiments with videos taken across several countries and cities, covering diverse urban scenarios, demonstrate the high scalability and generalizability of the proposed annotation method. Furthermore, performance analysis and real-world deployment for autonomous robot navigation showcase that the trained semantic traversability estimator is highly accurate, able to handle diverse camera viewpoints, computationally light, and real-world applicable. The summary video is available at https://youtu.be/EUVoH-wA-lA.

This paper focuses on transferring control policies between robot manipulators with different morphology. While reinforcement learning (RL) methods have shown successful results in robot manipulation tasks, transferring a trained policy from simulation to a real robot or deploying it on a robot with different states, actions, or kinematics is challenging. To achieve cross-embodiment policy transfer, our key insight is to project the state and action spaces of the source and target robots to a common latent space representation. We first introduce encoders and decoders to associate the states and actions of the source robot with a latent space. The encoders, decoders, and a latent space control policy are trained simultaneously using loss functions measuring task performance, latent dynamics consistency, and encoder-decoder ability to reconstruct the original states and actions. To transfer the learned control policy, we only need to train target encoders and decoders that align a new target domain to the latent space. We use generative adversarial training with cycle consistency and latent dynamics losses without access to the task reward or reward tuning in the target domain. We demonstrate sim-to-sim and sim-to-real manipulation policy transfer with source and target robots of different states, actions, and embodiments. The source code is available at \url{https://github.com/ExistentialRobotics/cross_embodiment_transfer}.

Therefore, to transfer a policy on the source robot to multiple target robots, they must launch multiple independent runs for each target robot. We investigate the problem of transferring an expert policy from a source robot to multiple different robots. To solve this problem, we propose a method named Meta-Evolve that uses continuous robot evolution to efficiently transfer the policy to each target robot through a set of tree-structured evolutionary robot sequences. The robot evolution tree allows the robot evolution paths to be shared, so our approach can significantly outperform naive one-to-one policy transfer. We present a heuristic approach to determine an optimized robot evolution tree. Experiments have shown that our method is able to improve the efficiency of one-to-three transfer of manipulation policy by up to 3.2 and one-to-six transfer of agile locomotion policy by 2.4 in terms of simulation cost over the baseline of launching multiple independent one-to-one policy transfers. The robotics industry has designed and developed a large number of commercial robots deployed in various applications. How to efficiently learn robotic skills on diverse robots in a scalable fashion? A popular solution is to train a policy for every new robot on every new task from scratch. This is not only inefficient in terms of sample efficiency but also impractical for complex robots due to a large exploration space. Inter-robot imitation by statistic matching methods that optimize to match the distribution of actions (Ross et al., 2011), transitioned states (Liu et al., 2019; Radosavovic et al., 2020), or reward (Ng et al., 2000; Ho & Ermon, 2016) could be possible solutions. However, they can only be applied to robots with similar dynamics to yield optimal performance. Recent advances in evolution-based imitation learning (Liu et al., 2022a;b) inspire us to view this problem from the perspective of policy transferring from one robot to another. The core idea is to interpolate two different robots by producing a large number of intermediate robots between them which gradually evolve from the source robot toward the target robot.

We propose a novel self-supervised approach for learning to visually localize robots equipped with controllable LEDs. We rely on a few training samples labeled with position ground truth and many training samples in which only the LED state is known, whose collection is cheap. We show that using LED state prediction as a pretext task significantly helps to learn the visual localization end task. The resulting model does not require knowledge of LED states during inference. We instantiate the approach to visual relative localization of nano-quadrotors: experimental results show that using our pretext task significantly improves localization accuracy (from 68.3% to 76.2%) and outperforms alternative strategies, such as a supervised baseline, model pre-training, and an autoencoding pretext task. We deploy our model aboard a 27-g Crazyflie nano-drone, running at 21 fps, in a position-tracking task of a peer nano-drone. Our approach, relying on position labels for only 300 images, yields a mean tracking error of 4.2 cm versus 11.9 cm of a supervised baseline model trained without our pretext task. Videos and code of the proposed approach are available at https://github.com/idsia-robotics/leds-as-pretext