This work proposes a hardware-friendly, dense optical flow-based Time-to-Collision (TTC) estimation algorithm intended to be deployed on smart video sensors for collision avoidance. The algorithm optimized for hardware first extracts biological visual motion features (motion energies), and then utilizes a Random Forests regressor to predict robust and dense optical flow. Finally, TTC is reliably estimated from the divergence of the optical flow field. This algorithm involves only feed-forward data flows with simple pixel-level operations, and hence has inherent parallelism for hardware acceleration. The algorithm offers good scalability, allowing for flexible tradeoffs among estimation accuracy, processing speed and hardware resource. Experimental evaluation shows that the accuracy of the optical flow estimation is improved due to the use of Random Forests compared to existing voting-based approaches. Furthermore, results show that estimated TTC values by the algorithm closely follow the ground truth. The specifics of the hardware design to implement the algorithm on a real-time embedded system are laid out.
In the field of industrial manufacturing, assembly line production is the most common production process that can be modeled as a permutation flow shop scheduling problem (PFSP). Minimizing the late work criteria (tasks remaining after due dates arrive) of production planning can effectively reduce production costs and allow for faster product delivery. In this article, a novel learning-based approach is proposed to minimize the late work of the PFSP using deep reinforcement learning (DRL) and graph isomorphism network (GIN), which is an innovative combination of the field of combinatorial optimization and deep learning. The PFSPs are the well-known permutation flow shop problem and each job comes with a release date constraint. In this work, the PFSP is defined as a Markov decision process (MDP) that can be solved by reinforcement learning (RL). A complete graph is introduced for describing the PFSP instance. The proposed policy network combines the graph representation of PFSP and the sequence information of jobs to predict the distribution of candidate jobs. The policy network will be invoked multiple times until a complete sequence is obtained. In order to further improve the quality of the solution obtained by reinforcement learning, an improved iterative greedy (IG) algorithm is proposed to search the solution locally. The experimental results show that the proposed RL and the combined method of RL+IG can obtain better solutions than other excellent heuristic and meta-heuristic algorithms in a short time.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.