Multi-Modal Model Predictive Control Through Batch Non-Holonomic Trajectory Optimization: Application to Highway Driving

Adajania, Vivek Kantilal; Sharma, Aditya; Gupta, Anish; Masnavi, Houman; Krishna, K. Madhava; Singh, Arun Kumar

doi:10.1109/lra.2022.3148460

Cited by 16 publications

(9 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Contribution over Author's Prior Work: Our current work adapts the batch optimization of [12] as a projection operator within a CEM-based gradient-free trajectory planning. This required non-trivial changes to the CEM pipeline and the batch optimizer update rule of [12]. Furthermore, we also expand the scope of [12] by including tracking constraints of the form (1d).…”

Section: B Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Visibility-Aware Navigation With Batch Projection Augmented Cross-Entropy Method Over a Learned Occlusion Cost

Masnavi

Shrestha

Mishra

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

We present two real-time trajectory optimizers based on the Cross-Entropy Method for visibility-aware navigation. The two approaches differ in handling inequality constraints stemming from bounds on motion derivatives, collision avoidance, tracking error, etc. Our first optimizer augments the inequalities into the cost function, while the second one relies on a novel GPU accelerated batch projection algorithm. We adopt a learningbased approach to ensure a fast query of the occlusion cost arising from the environment. Specifically, we train a neural network to compute the occlusion directly from the point obstacles generated from LiDAR or RGB-D sensors. Our learned occlusion model can be queried up to 3x faster than the approaches based on distance computation from occupancy or voxel maps. We improve the state-of-the-art in the following aspects. First, our optimizers do not require any explicit map building and can thus adapt on the fly to the changes in the environment. Second, we outperform existing approaches in target tracking applications in maintaining target visibility and success rate while being competitive in acceleration effort and computation time.

show abstract

Section: B Related Workmentioning

confidence: 99%

“…This required non-trivial changes to the CEM pipeline and the batch optimizer update rule of [12]. Furthermore, we also expand the scope of [12] by including tracking constraints of the form (1d).…”

Section: B Related Workmentioning

confidence: 99%

Visibility-Aware Navigation With Batch Projection Augmented Cross-Entropy Method Over a Learned Occlusion Cost

Masnavi

Shrestha

Mishra

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Each CPU core is efficient at handling arbitrary numerical computations, and thus solving a batch of optimizations problems in parallel is conceptually simple. We can solve each problem in a separate thread without needing to make any change in the underlying numerical algebra of the optimizer (Adajania et al, 2022), (Bento et al, 2013). As mentioned earlier, the scalability of CPU parallelization is limited by the number of cores (typical 6 in a standard laptop).…”

Section: Batch Optimization Over Cpu Vs Gpumentioning

confidence: 99%

“…The basic assumption is that the decoupled optimizations can be parallelized across separate CPU threads (Bento et al, 2013). However, our recent works have shown that such a parallelization approach does not scale well with an increase in the number of problems (Adajania et al, 2022). The inherent limitation stems from the available CPU cores and thread synchronization issues.…”

Section: Introductionmentioning

confidence: 99%

Fast Joint Multi-Robot Trajectory Optimization by GPU Accelerated Batch Solution of Distributed Sub-Problems

et al. 2022

View full text Add to dashboard Cite

We present a joint multi-robot trajectory optimizer that can compute trajectories for tens of robots in aerial swarms within a small fraction of a second. The computational efficiency of our approach is built on breaking the per-iteration computation of the joint optimization into smaller, decoupled sub-problems and solving them in parallel through a custom batch optimizer. We show that each of the sub-problems can be reformulated to have a special Quadratic Programming structure, wherein the matrices are shared across all the problems and only the associated vector varies. As result, the batch solution update rule reduces to computing just large matrix vector products which can be trivially accelerated using GPUs. We validate our optimizer’s performance in difficult benchmark scenarios and compare it against existing state-of-the-art approaches. We demonstrate remarkable improvements in computation time its scaling with respect to the number of robots. Moreover, we also perform better in trajectory quality as measured by smoothness and arc-length metrics.

show abstract

“…Second, the traffic distribution characteristics and influencing factors along the road section should be analyzed. Finally, the best design scheme should be selected in combination with the requirements of relevant specifications and the construction drawing budget should be completed to ensure the scientific and effective operation and safe maintenance of the highway [3,4].…”

Section: Introductionmentioning

confidence: 99%

Optimal Design of Mountain Expressway Based on Natural Environment Protection

2020

NEP

View full text Add to dashboard Cite

Optimizing the design of mountain roads can improve the service life of highways, reduce the maintenance costs in the later period, and promote the economic development of mountain areas. Based on the actual situation and data analysis of mountain expressway lines, this paper summarizes the natural ecological environment conditions and topographic and geomorphic features involved in the geological survey report, and discusses and studies the highway design according to the natural environment characteristics along the line. On the premise of ensuring the safety and reliability of the project, the ecological scheme is adopted to apply the optimal design scheme of mountain expressway, Under the condition that the base is stable and the landscape along the line is not strongly affected, the high dam scheme can be selected by increasing the base compactness, reducing waste and waste utilization, so as to reduce the project cost.

show abstract

Multi-Modal Model Predictive Control Through Batch Non-Holonomic Trajectory Optimization: Application to Highway Driving

Cited by 16 publications

References 15 publications

Visibility-Aware Navigation With Batch Projection Augmented Cross-Entropy Method Over a Learned Occlusion Cost

Visibility-Aware Navigation With Batch Projection Augmented Cross-Entropy Method Over a Learned Occlusion Cost

Fast Joint Multi-Robot Trajectory Optimization by GPU Accelerated Batch Solution of Distributed Sub-Problems

Optimal Design of Mountain Expressway Based on Natural Environment Protection

Contact Info

Product

Resources

About