Andrzej Reinke

Muhammad Fadhil Ginting, Dong-Ki Kim, Xiangyun Meng et al.

As robots become increasingly capable of operating over extended periods -- spanning days, weeks, and even months -- they are expected to accumulate knowledge of their environments and leverage this experience to assist humans more effectively. This paper studies the problem of Long-term Active Embodied Question Answering (LA-EQA), a new task in which a robot must both recall past experiences and actively explore its environment to answer complex, temporally-grounded questions. Unlike traditional EQA settings, which typically focus either on understanding the present environment alone or on recalling a single past observation, LA-EQA challenges an agent to reason over past, present, and possible future states, deciding when to explore, when to consult its memory, and when to stop gathering observations and provide a final answer. Standard EQA approaches based on large models struggle in this setting due to limited context windows, absence of persistent memory, and an inability to combine memory recall with active exploration. To address this, we propose a structured memory system for robots, inspired by the mind palace method from cognitive science. Our method encodes episodic experiences as scene-graph-based world instances, forming a reasoning and planning algorithm that enables targeted memory retrieval and guided navigation. To balance the exploration-recall trade-off, we introduce value-of-information-based stopping criteria that determines when the agent has gathered sufficient information. We evaluate our method on real-world experiments and introduce a new benchmark that spans popular simulation environments and actual industrial sites. Our approach significantly outperforms state-of-the-art baselines, yielding substantial gains in both answer accuracy and exploration efficiency.

Andrzej Reinke, Xieyuanli Chen, Cyrill Stachniss

Robust and reliable ego-motion is a key component of most autonomous mobile systems. Many odometry estimation methods have been developed using different sensors such as cameras or LiDARs. In this work, we present a resilient approach that exploits the redundancy of multiple odometry algorithms using a 3D LiDAR scanner and a monocular camera to provide reliable state estimation for autonomous vehicles. Our system utilizes a stack of odometry algorithms that run in parallel. It chooses from them the most promising pose estimation considering sanity checks using dynamic and kinematic constraints of the vehicle as well as a score computed between the current LiDAR scan and a locally built point cloud map. In this way, our method can exploit the advantages of different existing ego-motion estimating approaches. We evaluate our method on the KITTI Odometry dataset. The experimental results suggest that our approach is resilient to failure cases and achieves an overall better performance than individual odometry methods employed by our system.

Andrzej Reinke, Marco Camurri, Claudio Semini

Legged robots are becoming popular not only in research, but also in industry, where they can demonstrate their superiority over wheeled machines in a variety of applications. Either when acting as mobile manipulators or just as all-terrain ground vehicles, these machines need to precisely track the desired base and end-effector trajectories, perform Simultaneous Localization and Mapping (SLAM), and move in challenging environments, all while keeping balance. A crucial aspect for these tasks is that all onboard sensors must be properly calibrated and synchronized to provide consistent signals for all the software modules they feed. In this paper, we focus on the problem of calibrating the relative pose between a set of cameras and the base link of a quadruped robot. This pose is fundamental to successfully perform sensor fusion, state estimation, mapping, and any other task requiring visual feedback. To solve this problem, we propose an approach based on factor graphs that jointly optimizes the mutual position of the cameras and the robot base using kinematics and fiducial markers. We also quantitatively compare its performance with other state-of-the-art methods on the hydraulic quadruped robot HyQ. The proposed approach is simple, modular, and independent from external devices other than the fiducial marker.