Reinforcement Learning has emerged as a dominant post-training approach to elicit agentic RAG behaviors such as search and planning from language models.Despite its success with larger models, applying RL to compact models (e.g., 0.5-1B parameters) presents unique challenges.The compact models exhibit poor initial performance, resulting in sparse rewards and unstable training.To overcome these difficulties, we propose Distillation-Guided Policy Optimization (DGPO), which employs coldstart initialization from teacher demonstrations and continuous teacher guidance during policy optimization.To understand how compact models preserve agentic behavior, we introduce Agentic RAG Capabilities (ARCap), a fine-grained metric analyzing reasoning, search coordination, and response synthesis.Comprehensive experiments demonstrate that DGPO enables compact models to achieve sophisticated agentic search behaviors, even outperforming the larger teacher model in some cases.DGPO makes agentic RAG feasible in computing resource-constrained environments.

The human brain’s plasticity allows for the integration of artificial body parts into the human body. Leveraging this, embodied systems realize intuitive interactions with the environment. We introduce a novel concept: embodied swarm robots. Swarm robots constitute a collective of robots working in harmony to achieve a common objective, in our case, serving as functional body parts. Embodied swarm robots can dynamically alter their shape, density, and the correspondences between body parts and individual robots. We contribute an investigation of the influence on embodiment of swarm robot-specific factors derived from these characteristics, focusing on a hand. Our paper is the first to examine these factors through virtual reality (VR) and real-world robot studies to provide essential design considerations and applications of embodied swarm robots. Through quantitative and qualitative analysis, we identified a system configuration to achieve the embodiment of swarm robots.

The hierarchy of global and local planners is one of the most commonly utilized system designs in autonomous robot navigation. While the global planner generates a reference path from the current to goal locations based on the pre-built map, the local planner produces a kinodynamic trajectory to follow the reference path while avoiding perceived obstacles. To account for unforeseen or dynamic obstacles not present on the pre-built map, "when to replan" the reference path is critical for the success of safe and efficient navigation. However, determining the ideal timing to execute replanning in such partially unknown environments still remains an open question. In this work, we first conduct an extensive simulation experiment to compare several common replanning strategies and confirm that effective strategies are highly dependent on the environment as well as the global and local planners. Based on this insight, we then derive a new adaptive replanning strategy based on deep reinforcement learning, which can learn from experience to decide appropriate replanning timings in the given environment and planning setups. Our experimental results show that the proposed replanner can perform on par or even better than the current best-performing strategies in multiple situations regarding navigation robustness and efficiency.

Robot swarms hold immense potential for performing complex tasks far beyond the capabilities of individual robots. However, the challenge in unleashing this potential is the robots’ limited sensory capabilities, which hinder their ability to detect and adapt to unknown obstacles in real-time. To overcome this limitation, we introduce a novel robot swarm control method with an indirect obstacle detector using a smoothed particle hydrodynamics (SPH) model. The indirect obstacle detector can predict the collision with an obstacle and its collision point solely from the robot’s velocity information. This approach enables the swarm to effectively and accurately navigate environments without the need for explicit obstacle detection, significantly enhancing their operational robustness and efficiency. Our method’s superiority is quantitatively validated through a comparative analysis, showcasing its significant navigation and pattern formation improvements under obstacle-unaware conditions.

Due to the complex interactions between agents, learning multi-agent control policy often requires a prohibitive amount of data. This paper aims to enable multi-agent systems to effectively utilize past memories to adapt to novel collaborative tasks in a data-efficient fashion. We propose the Multi-Agent Coordination Skill Database, a repository for storing a collection of coordinated behaviors associated with key vectors distinctive to them. Our Transformer-based skill encoder effectively captures spatio-temporal interactions that contribute to coordination and provides a unique skill representation for each coordinated behavior. By leveraging only a small number of demonstrations of the target task, the database enables us to train the policy using a dataset augmented with the retrieved demonstrations. Experimental evaluations demonstrate that our method achieves a significantly higher success rate in push manipulation tasks compared with baseline methods like few-shot imitation learning. Furthermore, we validate the effectiveness of our retrieve-and-learn framework in a real environment using a team of wheeled robots.

Multi-agent path planning (MAPP) is the problem of planning collision-free trajectories from start to goal locations for a team of agents. This work explores a relatively unexplored setting of MAPP where streams of agents have to go through the starts and goals with high throughput. We tackle this problem by formulating a new variant of MAPP called periodic MAPP in which the timing of agent appearances is periodic. The objective with periodic MAPP is to find a periodic plan, a set of collision-free trajectories that the agent streams can use repeatedly over periods, with periods that are as small as possible. To meet this objective, we propose a solution method that is based on constraint relaxation and optimization. We show that the periodic plans once found can be used for a more practical case in which agents in a stream can appear at random times. We confirm the effectiveness of our method compared with baseline methods in terms of throughput in several scenarios that abstract autonomous intersection management tasks.

We present Neural A*, a novel data-driven search method for path planning problems. Despite the recent increasing attention to data-driven path planning, machine learning approaches to search-based planning are still challenging due to the discrete nature of search algorithms. In this work, we reformulate a canonical A* search algorithm to be differentiable and couple it with a convolutional encoder to form an end-to-end trainable neural network planner. Neural A* solves a path planning problem by encoding a problem instance to a guidance map and then performing the differentiable A* search with the guidance map. By learning to match the search results with ground-truth paths provided by experts, Neural A* can produce a path consistent with the ground truth accurately and efficiently. Our extensive experiments confirmed that Neural A* outperformed state-of-the-art data-driven planners in terms of the search optimality and efficiency trade-off. Furthermore, Neural A* successfully predicted realistic human trajectories by directly performing search-based planning on natural image inputs.

This paper proposes a retransmission control system based on image recognition accuracy as a traffic reduction technique for improving network bandwidth usage efficiency in image recognition service using wireless edge computing. For traffic reduction, image compression is useful. However, it is known to deteriorate the recognition accuracy. Our proposed system is applied to guarantee this deterioration. By retransmitting images according to the recognition accuracy, we aim to reduce traffic and to guarantee recognition accuracy. When compressing the image, PSNR is calculated to adaptively change the ratio of the image compression, to maintain the image quality. This paper uses down-sampler as an image compression method and demonstrates the effectiveness of the proposed system through a wireless network simulation. We confirm that our proposed retransmission system can reduce the network traffic congestion, and guarantee the recognition accuracy as same as the conventional method.

This work presents a deep reinforcement learning framework for interactive navigation in a crowded place. Our proposed approach, Learning to Balance (L2B) framework enables mobile robot agents to steer safely towards their destinations by avoiding collisions with a crowd, while actively clearing a path by asking nearby pedestrians to make room, if necessary, to keep their travel efficient. We observe that the safety and efficiency requirements in crowd-aware navigation have a trade-off in the presence of social dilemmas between the agent and the crowd. On the one hand, intervening in pedestrian paths too much to achieve instant efficiency will result in collapsing a natural crowd flow and may eventually put everyone, including the self, at risk of collisions. On the other hand, keeping in silence to avoid every single collision will lead to the agent's inefficient travel. With this observation, our L2B framework augments the reward function used in learning an interactive navigation policy to penalize frequent active path clearing and passive collision avoidance, which substantially improves the balance of the safety-efficiency trade-off. We evaluate our L2B framework in a challenging crowd simulation and demonstrate its superiority, in terms of both navigation success and collision rate, over a state-of-the-art navigation approach.

The lasso-based L 1 -graph is used in many applications since it can effectively model a set of data points as a graph. The lasso is a popular regression approach and the L 1 -graph represents data points as nodes by using the regression result. More specifically, by solving the L 1 -optimization problem of the lasso, the sparse regression coefficients are used to obtain the weights of the edges in the graph. Conventional graph structures such as k -NN graph use two steps, adjacency searching and weight selection, for constructing the graph whereas the lasso-based L 1 -graph derives the adjacency structure as well as the edge weights simultaneously by using a coordinate descent. However, the construction cost of the lasso-based L 1 -graph is impractical for large data sets since the coordinate descent iteratively updates the weights of all edges until convergence. Our proposal, Castnet, can efficiently construct the lasso-based L 1 -graph. In order to avoid updating the weights of all edges, we prune edges that cannot have nonzero weights before entering the iterations. In addition, we update edge weights only if they are nonzero in the iterations. Experiments show that Castnet is significantly faster than existing approaches.

This paper presents a new generalized (or ray-pixel, raxel) camera calibration algorithm for camera systems involving distortions by unknown refraction and reflection processes. The key idea is use of intersections of calibration planes, while conventional methods utilized collinearity constraints of points on the planes. We show that intersections of calibration planes can realize a simple linear algorithm, and that our method can be applied to any ray-distributions while conventional methods require knowing the ray-distribution class in advance. Evaluations using synthesized and real datasets demonstrate the performance of our method quantitatively and qualitatively.