Scheduling multiple tasks in a real-time, distributed fashion among multiple agents is indeed challenging. Scheduling becomes more challenging if task generation is dynamic in nature. This paper proposes an online real-time distributed dynamic task scheduling approach for multi-agent systems. The proposed scheduler is capable of rescheduling while the task generation is dynamic without hampering the current task execution, if any. The scalability of the proposed scheduler is realised by distributing all the agents into a number of groups. Distributed grouping makes the scheduler scalable for a large number of agents and tasks. Complexity analysis conveys that the time-complexity for scheduling is independent of agent count, but depends on the task count along with the agent’s payload. While the computational burden for the consensus formation varies linearly with the number of agents and is independent of task count. Simulation results confirm that the proposed scheduler outperforms the relevant reference scheduler in terms of successful consensus attained, average reward attained at consensus, task scheduled, task executed, dynamic scheduling, and scalability. One demonstration is shown in a typical warehouse environment.

In the field of reinforcement learning, reward functions are essential for training agents aiming to execute desired actions. However, designing reward functions manually is difficult, time-consuming, and can lead to inaccuracies. This research introduces a novel framework called the Scale In-variant Reward Framework (SIRF) for creating dense rewards using a neural network (NN) classifier. SIRF employs neural networks to mix complex state functions and generate dense bounded reward signals. SIRF's effectiveness is evaluated using the DDPG (Deep Deterministic Policy Gradient) algorithm in an OpenAI Gym environment, a popular open-source toolkit for reinforcement learning research. The results demonstrate that SIRF achieves comparable performance to existing reward functions from the OpenAI Gym environment in terms of key metrics such as average reward, network loss, velocity, and energy consumption for continuous environment agents. This suggests that SIRF can effectively guide the agent's learning process while maintaining consistency in the reward scale, simplifying the overall reinforcement learning setup.

This paper proposes a method for adapting an object-invariant action model for stacking operations. The proposed approach results in the generalization of stacking operations for multiple objects, in contrast to handcrafted action model-based task and motion planning (TAMP). Action model adaptation for new object stacking is done by drawing inferences using intuitions and learning a reward from visual clues based on the heuristically defined reward function. With this reward, the action model adaptation problem is formulated as an n-armed bandit problem. Additionally, an efficient action selection strategy is proposed for the n-armed bandit problem, which leads to fast convergence. The agent repetitively interacts with the environment in real-time to adapt the action model for the new object. After adaptation, the agent is capable of stacking objects with varying poses. The proposed approach is tested in simulation using the URS manipulator with a two-finger gripper. The proposed adaptive action model-based TAMP outperforms the traditional handcrafted action model-based TAMP by a significant margin in terms of success rate.

Recent advancements in legged locomotion research have made legged robots a preferred choice for navigating challenging terrains when compared to their wheeled counterparts. This paper presents a novel locomotion policy, trained using Deep Reinforcement Learning, for a quadrupedal robot equipped with an additional prismatic joint between the knee and foot of each leg. The training is performed in NVIDIA Isaac Gym simulation environment. Our study investigates the impact of these joints on maintaining the quadruped's desired height and following commanded velocities while traversing challenging terrains. We provide comparison results, based on a Cost of Transport (CoT) metric, between quadrupeds with and without prismatic joints. The learned policy is evaluated on a set of challenging terrains using the CoT metric in simulation. Our results demonstrate that the added degrees of actuation offer the locomotion policy more flexibility to use the extra joints to traverse terrains that would be deemed infeasible or prohibitively expensive for the conventional quadrupedal design, resulting in significantly improved efficiency.

This paper proposes a novel method to estimate the scale online for monocular SLAM. Unlike conventional scale estimation methods that require a metric sensor such as an IMU and apriori knowledge of its biases, this approach estimates the scale online by solely using the monocular camera and multirotor dynamics model in an extended Kalman Filter framework. Further, we discuss the observability of scale and theoretically show that the scale becomes observable when multirotor dynamics model and monocular vision are used in conjunction. We validate our proposition with extensive experimentation on the local as well as on the standard datasets and compare the same with other state of the art methods.

The paper proposes a path planning algorithm for cluttered environment and maze. The proposed planning algorithm exploits the merit of convex optimization while forming the convex navigable free-spaces, ensuring safety of the vehicle. The contiguous convex free-spaces are iteratively computed from a random-walk based seed generation method to create a contiguous navigable geometry. Inside this contiguous navigable geometry an undirected graph is then created, whose each node and edge belong to at least one convex region which boils down the path planning problem into a graph search problem. In addition, the proposed multiple query planning algorithm can merge the user provided feasible initial and goal configuration with the existing undirected graph in each plan, without deteriorating the planning performance in terms of run-time and path length. Simulation and experimental results confirm the superiority of the proposed planning algorithm jointly in terms of both path length and run-time by a significant margin.

This paper presents a novel method to eliminate drift for dynamics model based localization of multirotors. The dynamics equations require drag modelling, which is dependent on velocity, to generate vehicles's acceleration along the body axis. It is shown in this paper that the drag contribution, at velocity level, can be treated as a low frequency component. Incorrect or nonmodelling of this low frequency component leads to drift at velocity level. This drift can then be removed through a high pass filter. Experiments are performed on an AR.Drone quadrotor platform and results validated using motion capture camera system. The obtained results demonstrate that treating drag in the said manner leads to drift free velocity and hence to better localization. Further, it is shown that this approach could be exploited to extract scale information solely from a monocular camera.

With a view of robotic path planning inside any cluttered and confined indoor environment, a-priori generation of convex free-spaces offers additional benefit to the motion planner. Traditionally the convex free-space grows from a seed, which either selected randomly or generated from a heuristic function. However arbitrary seeding cannot guarantee maximal free-space coverage. In addition, improper seeding may ignore a valuable free-space information (e.g., narrow corridor, broken window, partially closed door etc.), because of which planning might fail. In the current work, a logically Extended Iterative Regional Inflation by Semidefinite Programming (EIRIS) is proposed to address the problem. A modified occupancy grip map (MOGM) creates an approximate binary map of the environment. It is thereby used to identify the mutually isolated largest possible free-regions. For each free-region proper seeding (PS) algorithm allows the generated seed to inflate along a coordinate axis. Unlike IRIS where unguided growth results in random free-space generation, the greedy expansion of a seed is inhibited temporarily by treating all the free-regions as virtual obstacles other than the region of interest. It is a guided iterative inflation (GIl) algorithm. Regarding robotic path planning the convex free-space information needs to be contiguous and subsequently traditional graph search technique can be exploited for efficient path generation. But it is beyond the scope of present discussion and considered as a future work. However few typical paths are shown in the paper to emphasize the utility of free-space generation in prior of path planning. Simulation results with empirical analysis establish the improvement of proposed method (EIRIS) over the state-of-the-art approach.

Numerous sensors are involved in sensing almost everything around us in a cyber-physical system. They generate heterogeneous data which are diverse in format and protocol. Prevailing standards formally define sensor data encodings and provide syntactic interoperability but failed to provide semantic capability required for searching, reasoning, querying on sensors and their observations. The paper extends IEEE SUMO and Onto Sensor to create an ontological structure that captures various sensor concepts and their relationships. It logically classifies sensor information into four major sensor knowledge representation (1) hierarchy (2) data (3) function and (4) data exchange. The proposed sensor ontology has been extended for energy meter as an energy sensor. The represented semantic can be captured through Protégé 4.3.0 and subsequently by OWLXML. Reasoning on the sensor data can be carried out using SWRL.