Actuator Control with Barrier Constraints for Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in responding immediately to obstacles due to complex calculations, making it difficult to avoid collisions with suddenly appearing or rapidly moving obstacles.
Innovation Solution
A controller that includes a setter, restrictor, and corrector to calculate and adjust actuator operation quantities, ensuring immediate obstacle avoidance by setting constraint conditions and correcting instruction values within a restricted range using control barrier functions and quadratic programming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If complex calculations are performed to generate optimal routes avoiding obstacles, then route optimization is improved, but calculation speed deteriorates
Solution Approach 1:
The control device segments the route generation process into two distinct phases: an offline phase where complex optimal route calculations are performed and stored as basic routes, and an online phase where only simple selection and execution of pre-calculated routes occur. This segmentation transfers the computational burden from real-time operation to offline preparation, resolving the contradiction between route optimization quality and calculation speed.
Solution Approach 2:
The system performs preliminary action by pre-calculating optimal routes offline before actual navigation begins. These pre-computed basic routes are stored and ready for immediate execution, eliminating the need for complex real-time calculations during obstacle avoidance maneuvers and enabling rapid response while maintaining route optimality.
2Reliability
If complex calculations are performed to avoid obstacles, then obstacle avoidance accuracy is improved, but response time deteriorates
Solution Approach 1:
The obstacle avoidance function is segmented into offline route planning (calculating optimal paths around anticipated obstacles) and online route selection (choosing from pre-calculated options when obstacles are detected). This allows the system to maintain high avoidance accuracy through thorough offline planning while achieving rapid response through simple online selection.
Solution Approach 2:
The system performs preliminary obstacle avoidance planning by pre-calculating multiple basic routes that account for potential obstacles. When obstacles are detected in real-time, the system simply selects from these pre-prepared options, ensuring both accurate avoidance and rapid response without performing complex calculations during the critical response period.
3Adaptability or versatility
If real-time route recalculation is performed for suddenly appearing obstacles, then obstacle avoidance capability is improved, but computational load increases
Solution Approach 1:
The system segments adaptability into two components: offline generation of multiple basic routes covering various obstacle scenarios, and online selection of the appropriate basic route based on current obstacle detection. This segmentation provides high adaptability to different obstacle situations while keeping online computational load minimal.
Solution Approach 2:
The system performs preliminary action by pre-generating multiple basic routes that anticipate various obstacle configurations. When sudden obstacles appear, the system adapts by selecting from these pre-computed routes rather than performing new calculations, thereby maintaining high adaptability while minimizing real-time computational requirements.
Data Source
AI summary
A controller 100 that calculates an operation quantity of an actuator 12 to move a mover 1 includes: a setter 22 that sets an instruction value to move the mover 1 to a target position; a restrictor 25 that sets a constraint condition of an instruction value for avoidance of an obstacle; a corrector 26 that corrects the instruction value to fall within a restricted range defined by the constraint condition if the instruction value is out of the restricted range; and an operation quantity calculator 28 that calculates the operation quantity based on the instruction value.